Faculty/ Deva Priyakumar U
Deva Priyakumar U
Professor
mail_outline
deva[at]iiit.ac.in
language
NEWS
Vishal Kumar - (Active)
vishal.kumar@research.iiit.ac.in
Sabarno Baral - (Active)
sabarno.b@research.iiit.ac.in
Athira Sasidharan Sasidharan - (Active)
athira.sasidharan@research.iiit.ac.in
NAVNEET BUNG - (Graduated)
b.navneet@research.iiit.ac.in
JAGANADE TANASHREE SANTOSH - (Graduated)
tanashree.jaganade@research.iiit.ac.in
Modee Rohit Laxman - (Graduated)
rohit.m@research.iiit.ac.in
Pradeep Kumar Pal - (Graduated)
pradeep.pal@research.iiit.ac.in
Suyash Gupta- (Active)
suyash.gupta@research.iiit.ac.in
Harsha Satya Vardhan Vasamsetti- (Active)
harsha.vasamsetti@research.iiit.ac.in
Manender Yadav- (Active)
manender.yadav@research.iiit.ac.in
Marisetti Sathya Abhijna- (Active)
sathya.marisetti@research.iiit.ac.in
Pratyusha Mitra- (Active)
pratyusha.mitra@research.iiit.ac.in
Polinati Jessy Grace- (Active)
jessygrace.polinati@research.iiit.ac.in
Khirade Gaurav Ramesh- (Active)
gaurav.kh@research.iiit.ac.in
Rishal Aggarwal- (Graduated)
rishal.aggarwal@research.iiit.ac.in
Abhiroop Talasila- (Graduated)
abhiroop.talasila@research.iiit.ac.in
Dhruv Rajeev Mittal- (Active)
dhruv.m@research.iiit.ac.in
Prakul Agrawal- (Active)
prakul.agrawal@research.iiit.ac.in
Arjun Dosajh- (Active)
arjun.dosajh@research.iiit.ac.in
Yashas S B- (Active)
yashas.b@research.iiit.ac.in
NITISH ALODIA- (Graduated)
nitish.alodia@research.iiit.ac.in
ACHIRA MALOO- (Graduated)
achira.maloo@research.iiit.ac.in
SAUMYA GURTU- (Graduated)
saumya.gurtu@research.iiit.ac.in
ADITYA CHATTOPADHYAY- (Graduated)
aditya.c@research.iiit.ac.in
SAGAR GAUR- (Graduated)
sagar.gaur@research.iiit.ac.in
PUNYASLOK PATTNAIK- (Graduated)
punyaslok.pattnaik@research.iiit.ac.in
MUSUNURU PRATYUSHA- (Graduated)
pratyusha.musunuru@research.iiit.ac.in
Shanmukh Alle- (Graduated)
shanmukh.alle@research.iiit.ac.in
Shreya Chandorkar- (Graduated)
shreya.chandorkar@research.iiit.ac.in
Vineeth Ravindra Chelur- (Graduated)
ravindrachelur.v@research.iiit.ac.in
Siddhartha Laghuvarapu- (Graduated)
siddhartha.l@research.iiit.ac.in
Yashaswi Pathak- (Graduated)
yashaswi.pathak@research.iiit.ac.in
Akshaya Karthikeyan- (Graduated)
akshaya.karthikeyan@research.iiit.ac.in
Sarvesh Mehta- (Graduated)
sarvesh.mehta@research.iiit.ac.in
Yashas Samaga B L- (Graduated)
yashas.samaga@research.iiit.ac.in
Manan Goel- (Graduated)
manan.goel@research.iiit.ac.in
S Bhuvanesh- (Graduated)
bhuvanesh.sridharan@research.iiit.ac.in
Akshit Garg- (Graduated)
akshit.garg@research.iiit.ac.in
V Vijay Vignesh- (Graduated)
vijay.vignesh@research.iiit.ac.in
Arihanth Srikar Tadanki- (Graduated)
arihanth.srikar@research.iiit.ac.in
Devata Sriram- (Graduated)
sriram.devata@research.iiit.ac.in
K Ganesh Chandan- (Graduated)
kanakala.ganesh@research.iiit.ac.in
Teaching Courses
SC2.203 - Biomolecular Structures
2025-26 - Spring
SC4.411 - Machine Learning for Natural Sciences
2025-26 - Spring
SC2.316 - Molecular Modeling and Simulations
2025-26 - Spring
SC4.101 - Computing in Sciences-I
2024-25 - Monsoon
SC2.203 - Biomolecular Structures
2024-25 - Spring
SC2.316 - Molecular Modeling and Simulations
2024-25 - Spring
SC4.101 - Computing in Sciences-I
2023-24 - Monsoon
SC2.203 - Biomolecular Structures
2023-24 - Spring
SC2.316 - Molecular Modeling and Simulations
2023-24 - Spring
SC4.412 - Data Driven Drug Discovery
2022-23 - Monsoon
SC2.316 - Molecular Modeling and Simulations
2022-23 - Spring
SC3.320 - Advanced Biomolecular Architecture
2021-22 - Monsoon
SC2.316 - Molecular Modeling and Simulations
2021-22 - Spring
SCI400 - CCNSB Seminar
2020-21 - Monsoon
SCI541 - Advanced Biomolecular Architecture
2020-21 - Monsoon
SCI400 - CCNSB Seminar
2020-21 - Spring
SC2.316 - Molecular Modeling and Simulations
2020-21 - Spring
SC4.120 - CCNSB Seminar - 1
2020-21 - Monsoon
SC9.600 - CCNSB Seminar-0 Credits
2020-21 - Spring
CN4.110 - Computing in Sciences-1
2019-20 - Monsoon
SCI373 - CNS Lab
2019-20 - Monsoon
SCI477 - Machine Learning for Natural Sciences
2019-20 - Monsoon
SC2.203 - Biomolecular Structures
2019-20 - Spring
SCI633 - Modeling and Simulations
2019-20 - Spring
SCI477 - Machine Learning for Natural Sciences
2019-20 - Spring
SCI373 - CNS Lab
2018-19 - Monsoon
CN4.110 - Computing in Sciences-1
2018-19 - Monsoon
SCI760 - Computer Aided Drug Design
2018-19 - Spring
SCI477 - Machine Learning for Natural Sciences
2018-19 - Spring
SCI331 - Thermodynamics & Statistical Mechanics
2017-18 - Monsoon
SCI373 - CNS Lab
2017-18 - Monsoon
SCI644 - Biomolecular Structure and Supramolecular Chemistry
2017-18 - Monsoon
SCI633 - Modeling and Simulations
2017-18 - Spring
SCI331 - Thermodynamics and Statistical Mechanics
2016-17 - Monsoon
SCI373 - CNS Lab
2016-17 - Monsoon
SCI644 - Biomolecular Structure and Supramolecular Chemistr
2016-17 - Monsoon
SCI633 - Modeling and Simulations
2016-17 - Spring
SCI760 - Computer Aided Drug Design
2016-17 - Spring
SCI760 - Computer Aided Drug Design
2016-17 - Spring
SCI633 - Modeling and Simulations
2015-16 - Spring
SCI644 - Biomolecular Structure and Supramolecular Chemistr
2015-16 - Monsoon
SCI760 - Computer Aided Drug Design
2014-15 - Spring
SCI341 - General and Structural Chemistry
2014-15 - Spring
About
Areas of Expertise
Selected Publications
Leveraging High-Spin DFT Features for Prediction of Spin State Gaps in 3d Transition Metal Complexes
Anuj Kumar Ray,Sandeep Kumar,Girish Varma,Deva Priyakumar U,Ankan Paul
Physical Chemistry Chemical Physics, PCCP, 2025
@inproceedings{bib_Leve_2025, AUTHOR = {Ray, Anuj Kumar and Kumar, Sandeep and Varma, Girish and U, Deva Priyakumar and Paul, Ankan }, TITLE = {Leveraging High-Spin DFT Features for Prediction of Spin State Gaps in 3d Transition Metal Complexes}, BOOKTITLE = {Physical Chemistry Chemical Physics}. YEAR = {2025}}
Determining spin state energy gaps (SSE) of 3d transition metal complexes (TMCs) is a major challenge in theoretical chemistry, as high-level quantum methods, though reliable, are computationally impractical for large-scale studies. This work explores a machine learning (ML)-based approach to predict DFT adiabatic SSE gaps using descriptors derived from a single high-spin DFT calculation. This approach is adopted to eliminate the differential treatment of electronic correlation between high-spin and low-spin structures. Our descriptors aim to incorporate the knowledge of crystal field theory into the ML model. They include atomic energy levels of bare metal ions, natural charges of ligating atoms, d-orbital molecular orbital eigenvalues derived from an high spin calculation, HOMO–LUMO gaps of free ligands, and simple identity-based features. We train ML models on 1434 SSE values spanning 934 complexes and demonstrate their transferability to more challenging complexes having bidentate π-bonding ligands despite being trained on simpler Werner-type monodentate complexes. We achieved a minimum MAE of 4.0 kcal mol−1 on the monodentate test set, and maintained a comparable MAE of 6.6 kcal mol−1 in the transferability assessment. This approach bypasses the need for multi-reference low-spin optimizations while retaining predictive accuracy, offering a cost-effective strategy for SSE estimation in transition metal chemistry. We hope the insights covered in this study will contribute to the development of additional electronic structure-based descriptors for SSE predictions.
Modern machine learning methods for protein property prediction
Arjun Dosajh,Prakul Agrawal,Prathit Chatterjee,Deva Priyakumar U
Current Opinion in Structural Biology, COBS, 2025
@inproceedings{bib_Mode_2025, AUTHOR = {Dosajh, Arjun and Agrawal, Prakul and Chatterjee, Prathit and U, Deva Priyakumar }, TITLE = {Modern machine learning methods for protein property prediction}, BOOKTITLE = {Current Opinion in Structural Biology}. YEAR = {2025}}
Recent progress and development of artificial intelligence and machine learning (AI/ML) techniques have enabled addressing complex biomolecular problems. AI/ML models learn the underlying distribution of data they are trained on and when exposed to new inputs, they make predictions based on patterns and relationships previously observed in the training set. Further, generative artificial intelligence (GenAI) can be used to accurately generate protein structure or sequence from specific selected properties. This review specifically focuses on the applications of AI/ML in predicting important functional properties of proteins, and the potential prospects of reverse-engineering in depicting the sequence and structure, from available protein-property information.
Kinetic vs Thermodynamic Controlled Promiscuous Biocatalytic Asymmetric Henry Reaction in Diastereocomplementary Synthesis of β-Nitroalcohols
Ayon Chatterjee,Pradeep Kumar Pal,Deva Priyakumar U,Santosh Kumar Padhi
ACS Catalysis, ACSC, 2025
@inproceedings{bib_Kine_2025, AUTHOR = {Chatterjee, Ayon and Pal, Pradeep Kumar and U, Deva Priyakumar and Padhi, Santosh Kumar }, TITLE = {Kinetic vs Thermodynamic Controlled Promiscuous Biocatalytic Asymmetric Henry Reaction in Diastereocomplementary Synthesis of β-Nitroalcohols}, BOOKTITLE = {ACS Catalysis}. YEAR = {2025}}
Exploring promiscuous catalytic activity of enzymes to stereoselective transformations uncovers ecofriendly asymmetric catalysts and provides scope to execute new-to-nature chemistry. Biocatalytic promiscuous diastereoselective Henry reaction (DHR) is restricted to a few hydroxynitrile lyases (HNLs), often engineered, with narrow synthetic scope and anti-selectivity. Here, we report a single native enzyme exhibiting complementary diastereoselectivity in an asymmetric Henry reaction. Baliospermum montanum HNL (BmHNL)-catalyzed DHR enabled the production of thirty-two anti-(1S,2R)-β-nitroalcohols with up to >99% conversion, >99% ee, and >99% de using longer nitroalkanes. It afforded syn diastereomers, i.e., (1S,2S)-β-nitroalcohols as thermodynamically controlled products by manipulating the reaction conditions. Reuse of BmHNL in DHR for 20 cycles has empowered the synthesis of (1S,2R)-2-nitro-1-phenylpropan-1-ol (NPP), where it retained 82% of its initial activity, maintained >99% ee on each cycle, and provided total turnover number >3.7 × 104, which is 140-fold higher compared to the only existing biocatalytic DHR. Our study demonstrates a gram-scale synthesis of (1S,2R)-NPP and a preparative-scale synthesis of (1S,2S)-NPP, a precursor to d-norpseudoephedrine used for the treatment of obesity. The origin of the inverse diastereoselectivity was probed using isotope labeling studies. Molecular modeling of the reaction using density functional theory methods reveals the underlying mechanism of the diastereoselectivity and the competing kinetic vs thermodynamic nature of the reaction. Despite a promiscuous catalytic activity, the outstanding catalytic efficiency, broad synthetic scope, and complementary diastereoselectivity of the native enzyme of BmHNL on DHR are remarkable, which illustrates it as a specialized biocatalyst for greener technology and stereocontrolled production of diverse β-nitroalcohols.
Theoretical Investigation of the Stabilities and Reactivities of AumCun Metallic Clusters (m+n = 13)
Pradeep Kumar Pal,Deva Priyakumar U
Chemistry - An Asian Journal, CAAJ, 2025
@inproceedings{bib_Theo_2025, AUTHOR = {Pal, Pradeep Kumar and U, Deva Priyakumar }, TITLE = {Theoretical Investigation of the Stabilities and Reactivities of AumCun Metallic Clusters (m+n = 13)}, BOOKTITLE = {Chemistry - An Asian Journal}. YEAR = {2025}}
AuCu nanoclusters have widespread application in reactions like activation of CO2, selective oxidation, and cross-coupling reactions. In this study, we investigate the stepwise doping of copper atoms in a pure 13-atom gold cluster, denoted as AumCun (m+n = 13). The genetic algorithm based on the artificial bee colony algorithm has been utilized to model various isomers of each composition. The potential energy landscape of these clusters was analyzed by means of the density functional theory method with pure Perdew−Burke−Ernzerhof (PBE) functional. We identify the minimum energy isomer for each cluster composition to evaluate molecular properties like HOMO–LUMO gap, binding energy/atom, second order difference in energy, vertical ionization energy, and vertical electron affinity. Notably, the introduction of copper atoms in these clusters enhances their stability and reactivity. Distinct odd–even oscillations due to close shell electronic configurations are absent, as all cluster compositions have an overall open shell configuration. To assess the catalytic activity of the clusters, we study the adsorption energies of small molecules like O 2 and C 2 H 4 on all available sites
on the cluster. This study thereby comprehensively explores the range of copper-doped 13-atom gold cluster compositions and their implications on their structure–property relationships vital for catalysis and nanomaterial applications.
Dissecting errors in machine learning for retrosynthesis: a granular metric framework and a transformer-based model for more informative predictions
Arihanth Srikar Tadanki,H Surya Prakash Rao,Deva Priyakumar U
Digital Discovery, DD, 2025
@inproceedings{bib_Diss_2025, AUTHOR = {Tadanki, Arihanth Srikar and Rao, H Surya Prakash and U, Deva Priyakumar }, TITLE = {Dissecting errors in machine learning for retrosynthesis: a granular metric framework and a transformer-based model for more informative predictions}, BOOKTITLE = {Digital Discovery}. YEAR = {2025}}
Chemical reaction prediction, encompassing forward synthesis and retrosynthesis, stands as a fundamental challenge in organic synthesis. A widely adopted computational approach frames synthesis prediction as a sequence-to-sequence translation task, using the commonly used SMILES representation for molecules. The current evaluation of machine learning methods for retrosynthesis assumes perfect training data, overlooking imperfections in reaction equations in popular datasets, such as missing reactants, products, other physical and practical constraints such as temperature and cost, primarily due to a focus on the target molecule. This limitation leads to an incomplete representation of viable synthetic routes, especially when multiple sets of reactants can yield a given desired product. In response to these shortcomings, this study examines the prevailing evaluation methods and introduces comprehensive metrics designed to address imperfections in the dataset. Our novel metrics not only assess absolute accuracy by comparing predicted outputs with ground truth but also introduce a nuanced evaluation approach. We provide scores for partial correctness and compute adjusted accuracy through graph matching, acknowledging the inherent complexities of retrosynthetic pathways. Additionally, we explore the impact of small molecular augmentations while preserving chemical properties and employ similarity matching to enhance the assessment of prediction quality. We introduce SynFormer, a sequence-to-sequence model tailored for SMILES representation. It incorporates architectural enhancements to the original transformer, effectively tackling the challenges of chemical reaction prediction. SynFormer achieves a Top-1 accuracy of 53.2% on the USPTO-50k dataset, matching the performance of widely accepted models like Chemformer, but with greater efficiency by eliminating the need for pre-training.
GraphDDI: Graph Neural Network for Prediction of Drug-Drug Interaction
Suyash Gupta,Deva Priyakumar U,Siddhartha Laghuvarapu
Conference on AI in Healthcare, ICAIHC, 2024
Abs | | bib Tex
@inproceedings{bib_Grap_2024, AUTHOR = {Gupta, Suyash and U, Deva Priyakumar and Laghuvarapu, Siddhartha }, TITLE = {GraphDDI: Graph Neural Network for Prediction of Drug-Drug Interaction}, BOOKTITLE = {Conference on AI in Healthcare}. YEAR = {2024}}
Drug-Drug Interactions (DDI) can trigger unexpected pharmacological consequences, including adverse drug events (ADE). The rise in polypharmacy underscores the importance of understanding how different drug molecules influence each other’s pharmacological activities and necessitates the investigation of potential interactions between newly developed drugs and existing medications. Traditional laboratory methods for DDI detection are time-consuming, making the development of computational prediction methods crucial. This study introduces GraphDDI, a machine learning method utilizing Graph Neural Networks (GNN) to predict DDI accurately. The proposed methodology, trained end-to-end, comprises three stages. (1) Featurization stage: A GNN extracts atomic features from two drugs separately. (2) Interaction stage: an interaction map is calculated between all atom pairs of the drugs. (3) Prediction stage: the model combines the interaction map and drug features to create a unified representation of the drug molecules. Subsequently, the model concatenates these representations and employs a feed-forward neural network to predict the DDI. We demonstrate the efficacy of our proposed model in predicting both the presence of DDI and the specific types of interactions (DDI events). Comparative analysis reveals that our framework surpasses existing models, achieving an F1 score of 0.98 in predicting the existence of drug-drug interactions and 0.90 in categorizing DDI event types. The code is available in our GitHub repository (https://github.com/devalab/GraphDDI).
Generative artificial intelligence for small molecule drug design
Ganesh Chandan Kanakala,Sriram Devata,Prathit Chatterjee,Deva Priyakumar U
Current Opinion in Biotechnology, COBT, 2024
Abs | | bib Tex
@inproceedings{bib_Gene_2024, AUTHOR = {Kanakala, Ganesh Chandan and Devata, Sriram and Chatterjee, Prathit and U, Deva Priyakumar }, TITLE = {Generative artificial intelligence for small molecule drug design}, BOOKTITLE = {Current Opinion in Biotechnology}. YEAR = {2024}}
In recent years, the rapid advancement of generative artificial intelligence (GenAI) has revolutionized the landscape of drug design, offering innovative solutions to potentially expedite the discovery of novel therapeutics. GenAI encompasses algorithms and models that autonomously create new data, including text, images, and molecules, often mirroring characteristics of existing datasets. This comprehensive review delves into the realm of GenAI for drug design, emphasizing recent advancements and methodologies that have propelled the field forward. Specifically, we focus on three prominent paradigms: transformers, diffusion models, and reinforcement learning algorithms, which have been exceptionally impactful in the last few years. By synthesizing insights from a myriad of studies and developments, we elucidate the potential of these approaches in accelerating the drug discovery process. Through a detailed analysis, we explore the current state and future directions of GenAI in the context of drug design, highlighting its transformative impact on pharmaceutical research and development.
Dissecting Errors in Machine Learning for Retrosynthesis: A Granular Metric Framework and Transformer-Based Model for More Informative Predictions
Arihanth Srikar Tadanki,H. Surya Prakash Rao,Deva Priyakumar U
Digital Discovery, DD, 2024
@inproceedings{bib_Diss_2024, AUTHOR = {Tadanki, Arihanth Srikar and Rao, H. Surya Prakash and U, Deva Priyakumar }, TITLE = {Dissecting Errors in Machine Learning for Retrosynthesis: A Granular Metric Framework and Transformer-Based Model for More Informative Predictions}, BOOKTITLE = {Digital Discovery}. YEAR = {2024}}
Chemical reaction prediction, encompassing forward synthesis and retrosynthesis, stands as a fundamental challenge in organic synthesis. A widely adopted computational approach frames synthesis prediction as a sequence-to-sequence translation task, using the common SMILES representation for molecules. Current evaluation of machine learning methods for retrosynthesis assume perfect training data, overlooking imperfections in reaction equations in popular datasets, such as missing reactants, products, other physical and practical constraints such as temperature and cost, primarily driven by a focus on the target molecule. This limitation leads to an incomplete representation of viable synthetic routes, especially when multiple sets of reactants can yield a given desired product. In response to these shortcomings, this study examines the prevailing evaluation methods and introduces comprehensive metrics designed to address imperfections in the dataset. Our novel metrics not only assess absolute accuracy by comparing predicted outputs with ground truth but also introduce a nuanced evaluation approach. We provide scores for partial correctness and compute adjusted accuracy through graph matching, acknowledging the inherent complexities of retrosynthetic pathways. Additionally, we explore the impact of small molecular augmentations while preserving chemical properties and employ similarity matching to enhance the assessment of prediction quality. We introduce SynFormer, a sequence-to-sequence model tailored for SMILES representation. It incorporates architectural enhancements to the original transformer, effectively tackling the challenges of chemical reaction prediction. SynFormer achieves a top-1 accuracy of 53.2% on the USPTO-50k dataset, demonstrating an improvement over previous state-of-the-art language models while being more efficient and eliminating the need for pre-training.
TorRNA - Improved Prediction of Backbone Torsion Angles of RNA by Leveraging Large Language Models
Devata Sriram,Deva Priyakumar U
Chem arxiv, Chem arxiv, 2024
@inproceedings{bib_TorR_2024, AUTHOR = {Sriram, Devata and U, Deva Priyakumar }, TITLE = {TorRNA - Improved Prediction of Backbone Torsion Angles of RNA by Leveraging Large Language Models}, BOOKTITLE = {Chem arxiv}. YEAR = {2024}}
RNA molecules play a significant role in many biological pathways and have diverse
functional roles, which is a result of their structural flexibility to fold into diverse conformations. This structural flexibility makes it challenging to obtain the structures of
RNAs experimentally. Deep learning can be used to predict the secondary structures of
RNA and other properties such as the backbone torsion angles, to be used as restraints
for the computational optimization of the tertiary structures of RNA. TorRNA is a
transformer encoder-decoder model, that takes an input RNA sequence and predicts
the (pseudo)torsion angles of each nucleotide with a pre-trained RNA-FM model as
the encoder. TorRNA is able to achieve a performance boost of 2% − 16% over the
previous (pseudo)torsion angle prediction method for RNAs. We also demonstrate that
TorRNA can used as a tool for model quality assessment of candidate RNA structures
PLAS-20k: Extended Dataset of Protein-Ligand Affinities from MD Simulations for Machine Learning Applications
Saalim H. Raza,Shubham Sharma,Shruti S Jeurkar,Kavita Thakran,Reena Jaglan,Shivangi Verma,Deva Priyakumar U,Divya Nayar,Prathit Chatterjee,Indhu Ramachandran,Divya B Kolepara,Vasavi C.S.,Rakesh Srivastava,Pradeep Kumar Pal,Vishal Kumar,Shivam Pandit,Aathira G. Nair,Sanjana Pandey
Scientific Data, SD, 2024
@inproceedings{bib_PLAS_2024, AUTHOR = {Raza, Saalim H. and Sharma, Shubham and Jeurkar, Shruti S and Thakran, Kavita and Jaglan, Reena and Verma, Shivangi and U, Deva Priyakumar and Nayar, Divya and Chatterjee, Prathit and Ramachandran, Indhu and Kolepara, Divya B and C.S., Vasavi and Srivastava, Rakesh and Pal, Pradeep Kumar and Kumar, Vishal and Pandit, Shivam and Nair, Aathira G. and Pandey, Sanjana }, TITLE = {PLAS-20k: Extended Dataset of Protein-Ligand Affinities from MD Simulations for Machine Learning Applications}, BOOKTITLE = {Scientific Data}. YEAR = {2024}}
Computing binding affinities is of great importance in drug discovery pipeline and its prediction using advanced machine learning methods still remains a major challenge as the existing datasets and models do not consider the dynamic features of protein-ligand interactions. To this end, we have developed PLAS-20k dataset, an extension of previously developed PLAS-5k, with 97,500 independent simulations on a total of 19,500 different protein-ligand complexes. Our results show good correlation with the available experimental values, performing better than docking scores. This holds true even for a subset of ligands that follows Lipinski’s rule, and for diverse clusters of complex structures, thereby highlighting the importance of PLAS-20k dataset in developing new ML models. Along with this, our dataset is also beneficial in classifying strong and weak binders compared to docking. Further, OnionNet model has been retrained on PLAS-20k dataset and is provided as a baseline for the prediction of binding affinities. We believe that large-scale MD-based datasets along with trajectories will form new synergy, paving the way for accelerating drug discovery.
DeepSPInN - Deep reinforcement learning for molecular Structure Prediction from Infrared and 13C NMR spectra
Devata Sriram,S Bhuvanesh,Sarvesh Mehta,Yashaswi Pathak,Siddhartha Laghuvarapu,Girish Varma,Deva Priyakumar U
Digital Discovery, DD, 2024
Abs | | bib Tex
@inproceedings{bib_Deep_2024, AUTHOR = {Sriram, Devata and Bhuvanesh, S and Mehta, Sarvesh and Pathak, Yashaswi and Laghuvarapu, Siddhartha and Varma, Girish and U, Deva Priyakumar }, TITLE = {DeepSPInN - Deep reinforcement learning for molecular Structure Prediction from Infrared and 13C NMR spectra}, BOOKTITLE = {Digital Discovery}. YEAR = {2024}}
Molecular spectroscopy studies the interaction of molecules with electromagnetic radiation, and interpreting the resultant spectra is invaluable for deducing the molecular structures. However, predicting the molecular structure from spectroscopic data is a strenuous task that requires highly specific domain knowledge. DeepSPInN is a deep reinforcement learning method that predicts the molecular structure when given Infrared and 13C Nuclear magnetic resonance spectra by formulating the molecular structure prediction problem as a Markov decision process (MDP) and employs Monte-Carlo tree search to explore and choose the actions in the formulated MDP. On the QM9 dataset, DeepSPInN is able to predict the correct molecular structure for 91.5% of the input spectra in an average time of 77 seconds for molecules with less than 10 heavy atoms. This study is the first of its kind that uses only infrared and 13C nuclear magnetic resonance spectra for molecular structure prediction without referring to any pre-existing spectral databases or molecular fragment knowledge bases, and is a leap forward in automated molecular spectral analysis.
Molecular Property Diagnostic Suite for COVID-19 (MPDSCOVID-19): An open access disease specific drug discovery porta
Deva Priyakumar U
bioRxiv, bioRxiv, 2023
@inproceedings{bib_Mole_2023, AUTHOR = {U, Deva Priyakumar }, TITLE = {Molecular Property Diagnostic Suite for COVID-19 (MPDSCOVID-19): An
open access disease specific drug discovery porta}, BOOKTITLE = {bioRxiv}. YEAR = {2023}}
Computational drug discovery is intrinsically interdisciplinary and has to deal with the multifarious factors which are often dependent on the type of disease. Molecular Property Diagnostic Suite (MPDS) is a Galaxy based web portal which was conceived and developed as a disease specific web portal, originally developed for tuberculosis (MPDSTB). As specific computational tools are often required for a given disease, developing a disease specific web portal is highly desirable. This paper emphasises on the development of the customised web portal for COVID-19 infection and is referred to as MPDSCOVID-19. Expectedly, the MPDS suites of programs have modules which are essentially independent of a given disease, whereas some modules are specific to a particular disease. In the MPDSCOVID-19 portal, there are modules which are specific to COVID-19, and these are clubbed in SARS-COV-2 disease library. Further, the new additions and/or significant improvements were made to the disease independent modules, besides the addition of tools from galaxy toolshed. This manuscript provides a latest update on the disease independent modules of MPDS after almost 6 years, as well as provide the contemporary information and tool-shed necessary to engage in the drug discovery research of COVID-19. The disease independent modules include file format converter and descriptor calculation under the data processing module; QSAR, pharmacophore, scaffold analysis, active site analysis, docking, screening, drug repurposing tool, virtual screening, visualisation, sequence alignment, phylogenetic analysis under the data analysis module; and various machine learning packages, algorithms and in-house developed machine learning antiviral prediction model are available. The MPDS suite of programs are expected to bring a paradigm shift in computational drug discovery, especially in the academic community, guided through a transparent and open innovation approach. The MPDSCOVID-19 can be accessed at http://mpds.neist.res.in:8085.
Single-Lead to Multi-Lead Electrocardiogram Reconstruction Using a Modified Attention U-Net Framework
Akshit Garg,V Vijay Vignesh,Deva Priyakumar U
International Joint Conference on Neural Networks, IJCNN, 2023
@inproceedings{bib_Sing_2023, AUTHOR = {Garg, Akshit and Vignesh, V Vijay and U, Deva Priyakumar }, TITLE = {Single-Lead to Multi-Lead Electrocardiogram
Reconstruction Using a Modified Attention U-Net
Framework}, BOOKTITLE = {International Joint Conference on Neural Networks}. YEAR = {2023}}
With Cardiovascular Diseases on the rise around
the world, Electrocardiograms (ECGs) play a crucial role in
their diagnosis owing to their non-invasive nature and simplicity.
Medical professionals typically use 12-lead ECGs for medical
analysis but gathering 12-lead ECG data is an arduous task
outside clinical setting. Modern wearables can collect an ECG
with fewer leads than the standard 12 leads. However, medical
professionals and conventional ECG analysis software find this
reduced lead set data challenging to interpret. By using the
reduced lead set data to create standard 12-lead ECG data,
ECG reconstruction can solve this issue. This paper proposes
a novel single-lead to multi-lead ECG reconstruction solution
using a modified Attention U-net framework. Using only the
lead II of ECG, our model is able to reproduce the other 11
leads of conventional 12-lead ECG with a Pearson correlation,
Mean square error and R-squared value of 0.805, 0.0122 and
0.639, respectively. Further, a single combined model is used to
reconstruct all 11 leads simultaneously, improving performance
and simultaneously reducing the computational resources needed
for training compared to current literature in the field. In
comparison to previous works, which only reconstruct small ECG
segments, our model is trained to reconstruct longer 10-second
ECG signals. We demonstrate our model’s ability for real-life
utilisation using a cardiovascular disease classification task. A
deep learning model was trained for multi-disease classification
on actual 12-lead ECG data and was tested on both original and
reconstructed 12-lead ECG signals. The classification accuracies
for the original and reconstructed signals were comparable,
portraying that our reconstruction model can preserve diagnostically relevant artefacts in its reconstructed signals. This work
provides a new promising solution in the field of single-lead ECG
reconstruction, taking us a step closer to bridging the divide
between reduced lead set data and existing 12-lead ECG end
users like clinicians and automatic ECG classifiers.
A Machine Learning Approach for Outcome Prediction in Postanoxic Coma Patients Using Frequency Domain Features
V Vijay Vignesh,Akshit Garg,Maitreya Maity,Deva Priyakumar U
Computing in Cardiology, CinC, 2023
@inproceedings{bib_A_Ma_2023, AUTHOR = {Vignesh, V Vijay and Garg, Akshit and Maity, Maitreya and U, Deva Priyakumar }, TITLE = {A Machine Learning Approach for Outcome Prediction in Postanoxic Coma Patients Using Frequency Domain Features}, BOOKTITLE = {Computing in Cardiology}. YEAR = {2023}}
In this work, we describe the creation of our machine- learning-based solution for coma prognosis after cardiac arrest using longitudinal EEG and ECG recordings for the ”Predicting Neurological Recovery from Coma After Car- diac Arrest: The George B. Moody PhysioNet Challenge 2023”. Our team, “ComaToast”, had its best submission ranked 28 out of 36 teams selected worldwide, with a chal- lenge score of 0.381 on the official leaderboard for the hid- den test set. We use a combination of age and signal fea- tures from EEG and ECG recordings. Frequency domain features, specifically mean power spectral density from 4 different bands of frequencies (Delta, Theta, Alpha and Beta) and mean Burst Suppression Ratio, were extracted from pre-processed EEG recordings from the first and last available recording for a given patient. Features like mean and standard deviations were extracted along channels for ECG recordings. After imputing missing values, these fea- tures are fed to an XGBoost classifier for the final binary classification of the outcome prediction task. The features are fed to a random forest regressor to predict the CPC outcome for every patient. A solution like ours, which uses a simple model and training technique, may be more viable than deep-learning solutions in general use cases. In our final model, our approach achieved a 5-fold cross- validation score of 0.34 on the public train set.
Streamlining pipeline efficiency: a novel model-agnostic technique for accelerating conditional generative and virtual screening pipelines
Karthik Viswanathan,Manan Goel,Siddhartha Laghuvarapu,Girish Varma,Deva Priyakumar U
NPG Nature Scientific Reports, NPG, 2023
@inproceedings{bib_Stre_2023, AUTHOR = {Viswanathan, Karthik and Goel, Manan and Laghuvarapu, Siddhartha and Varma, Girish and U, Deva Priyakumar }, TITLE = {Streamlining pipeline efficiency: a novel model-agnostic technique for accelerating conditional generative and virtual screening pipelines}, BOOKTITLE = {NPG Nature Scientific Reports}. YEAR = {2023}}
The discovery of potential therapeutic agents for life‑threatening diseases has become a significant problem. There is a requirement for fast and accurate methods to identify drug‑like molecules that can be used as potential candidates for novel targets. Existing techniques like high‑throughput screening and virtual screening are time‑consuming and inefficient. Traditional molecule generation pipelines are more efficient than virtual screening but use time‑consuming docking software. Such docking functions can be emulated using Machine Learning models with comparable accuracy and faster execution times. However, we find that when pre‑trained machine learning models are employed in generative pipelines as oracles, they suffer from model degradation in areas where data is scarce. In this study, we propose an active learning‑based model that can be added as a supplement to enhanced molecule generation architectures. The proposed method uses uncertainty sampling on the molecules created by the generator model and dynamically learns as the generator samples molecules from different regions of the chemical space. The proposed framework can generate molecules with high binding affinity with ∼a 70% improvement in runtime compared to the baseline model by labeling only ∼30% of molecules compared to the baseline oracle.
MolOpt: Autonomous Molecular Geometry Optimization Using Multiagent Reinforcement Learning
Modee Rohit Laxman,Sarvesh Mehta,Siddhartha Laghuvarapu,Deva Priyakumar U
The Journal of Physical Chemistry B, JPCB, 2023
@inproceedings{bib_MolO_2023, AUTHOR = {Laxman, Modee Rohit and Mehta, Sarvesh and Laghuvarapu, Siddhartha and U, Deva Priyakumar }, TITLE = {MolOpt: Autonomous Molecular Geometry Optimization Using Multiagent Reinforcement Learning}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2023}}
In this paper, we propose MolOpt, the first attempt of its kind to use Multi-Agent Reinforcement Learning (MARL) for autonomous molecular geometry optimization (MGO). Typically MGO algorithms are hand-designed, but MolOpt uses MARL to learn a learned optimizer (policy) that can perform MGO without depending on other hand-designed optimizers. We cast MGO as a MARL problem, where each agent corresponds to a single atom in the molecule. MolOpt performs MGO by minimizing the forces on each atom in the molecule. Our experiments demonstrate the generalizing ability of MolOpt for MGO of Propane, Pentane, Heptane, Hexane, and Octane when trained on Ethane, Butane, and Isobutane. In terms of performance, MolOpt outperforms the MDMin optimizer and demonstrates similar performance to the FIRE optimizer. However, it does not surpass the BFGS optimizer. The results demonstrate that MolOpt has the potential to introduce innovative advancements in MGO by providing a novel approach using reinforcement learning (RL), which may open up new research directions for MGO. Overall, this work serves as a proof-of-concept for the potential of MARL in MGO.
Streamlining pipeline efficiency: a novel model-agnostic technique for accelerating conditional generative and virtual screening pipelines
Viswanath Kasturi,Manan Goel,Siddhartha Laghuvarapu,Girish Varma,Deva Priyakumar U
Scientific Reports, SR, 2023
@inproceedings{bib_Stre_2023, AUTHOR = {Kasturi, Viswanath and Goel, Manan and Laghuvarapu, Siddhartha and Varma, Girish and U, Deva Priyakumar }, TITLE = {Streamlining pipeline efficiency: a novel model-agnostic technique for accelerating conditional generative and virtual screening pipelines}, BOOKTITLE = {Scientific Reports}. YEAR = {2023}}
The discovery of potential therapeutic agents for life-threatening diseases has become a significant problem. There is a requirement for fast and accurate methods to identify drug-like molecules that can be used as potential candidates for novel targets. Existing techniques like high-throughput screening and virtual screening are time-consuming and inefficient. Traditional molecule generation pipelines are more efficient than virtual screening but use time-consuming docking software. Such docking functions can be emulated using Machine Learning models with comparable accuracy and faster execution times. However, we find that when pre-trained machine learning models are employed in generative pipelines as oracles, they suffer from model degradation in areas where data is scarce. In this study, we propose an active learning-based model that can be added as a supplement to enhanced molecule generation architectures. The proposed method uses uncertainty sampling on the molecules created by the generator model and dynamically learns as the generator samples molecules from different regions of the chemical space. The proposed framework can generate molecules with high binding affinity with ∼ a 70% improvement in runtime compared to the baseline model by labeling only ∼ 30% of molecules compared to the baseline oracle.
Efficient and enhanced sampling of drug-like chemical space for virtual screening and molecular design using modern machine learning methods
Sarvesh Mehta,Siddhartha Laghuvarapu,Yashaswi Pathak,Aaftaab Sethi,Mallika Alvala,Deva Priyakumar U
Chem arxiv, Chem arxiv, 2023
@inproceedings{bib_Effi_2023, AUTHOR = {Mehta, Sarvesh and Laghuvarapu, Siddhartha and Pathak, Yashaswi and Sethi, Aaftaab and Alvala, Mallika and U, Deva Priyakumar }, TITLE = {Efficient and enhanced sampling of drug-like chemical space for virtual screening and molecular design using modern machine learning methods}, BOOKTITLE = {Chem arxiv}. YEAR = {2023}}
In drug discovery applications, high throughput virtual screening exercises are rou- tinely performed to determine an initial set of candidate molecules referred to as “hits”. In such an experiment, each molecule from large small-molecule drug library is evalu- ated for physical property such as the binding affinity (docking score) against a target receptor. In real-life drug discovery experiments, the drug libraries are extremely large but still a minor representation of the essentially infinite chemical space , and evaluation
Latent Biases in Machine Learning Models for Predicting Binding Affinities Using Popular Data Sets
K Ganesh Chandan,Rishal Aggarwal,Divya Nayar,Deva Priyakumar U
ACS Omega, OMEGA, 2023
@inproceedings{bib_Late_2023, AUTHOR = {Chandan, K Ganesh and Aggarwal, Rishal and Nayar, Divya and U, Deva Priyakumar }, TITLE = {Latent Biases in Machine Learning Models for Predicting Binding Affinities Using Popular Data Sets}, BOOKTITLE = {ACS Omega}. YEAR = {2023}}
Drug design involves the process of identifying and designing molecules that bind well to a given receptor. A vital computational component of this process is the protein–ligand interaction scoring functions that evaluate the binding ability of various molecules or ligands with a given protein receptor binding pocket reasonably accurately. With the publicly available protein–ligand binding affinity data sets in both sequential and structural forms, machine learning methods have gained traction as a top choice for developing such scoring functions. While the performance shown by these models is optimistic, there are several hidden biases present in these data sets themselves that affect the utility of such models for practical purposes such as virtual screening. In this work, we use published methods to systematically investigate several such factors or biases present in these data sets. In our analysis, we highlight the importance of considering sequence, protein–ligand interaction, and pocket structure similarity while constructing data splits and provide an explanation for good protein-only and ligand-only performances in some data sets. Through this study, we provide to the community several pointers for the design of binding affinity predictors and data sets for reliable applicability.
Self-Supervision and Weak Supervision for Accurate and Interpretable Chest X-Ray Classification Models
Abhiroop Talasila,Akshaya Karthikeyan,Shanmukh Alle,Maitreya Maity,Deva Priyakumar U
International Joint Conference on Neural Networks, IJCNN, 2023
Abs | | bib Tex
@inproceedings{bib_Self_2023, AUTHOR = {Talasila, Abhiroop and Karthikeyan, Akshaya and Alle, Shanmukh and Maity, Maitreya and U, Deva Priyakumar }, TITLE = {Self-Supervision and Weak Supervision for Accurate and Interpretable Chest X-Ray Classification Models}, BOOKTITLE = {International Joint Conference on Neural Networks}. YEAR = {2023}}
X-rays diagnose numerous thoracic diseases; however, accurate pathology detection requires trained radiologists. The number of available experts may impede population-level patient care, delaying medical action. State-of-the-art (SOTA) machine learning methods categorize chest X-rays across numerous diseases well but do not always account for explainability. Interpretability assessments rarely focus on the conciseness and anatomical correctness of Class Activation Mapping outcomes. These models are not accurate or dependable when tested on different datasets of the same modality. They are not used because their predicted performance is not explainable. This work introduces a self-supervised and weakly supervised pretraining pipeline with an auxiliary loss and supervised fine-tuning that retains performance across datasets. We use the Chest X-ray14 (NIH CXR) dataset for pre-training and CheXpert
PLAS-20k: Extended Dataset of Protein-Ligand Affinities from MD Simulations for Machine Learning Applications
Divya B Kolepara,Shubham Sharma,Shruti S Jeurkar,Kavita Thakran,Reena Jaglan,Shivangi Verma,Deva Priyakumar U,C. S. Vasavi1,Rakesh Srivastava,Pradeep Kumar Pal,Saalim H. Raza,Vishal Kumar, Shivam Pandit,Aathira G Nair,Sanjana Pandey
Chem arxiv, Chem arxiv, 2023
@inproceedings{bib_PLAS_2023, AUTHOR = {Kolepara, Divya B and Sharma, Shubham and Jeurkar, Shruti S and Thakran, Kavita and Jaglan, Reena and Verma, Shivangi and U, Deva Priyakumar and Vasavi1, C. S. and Srivastava, Rakesh and Pal, Pradeep Kumar and Raza, Saalim H. and Kumar, Vishal and Pandit, Shivam and Nair, Aathira G and Pandey, Sanjana }, TITLE = {PLAS-20k: Extended Dataset of Protein-Ligand Affinities from MD Simulations for Machine Learning Applications}, BOOKTITLE = {Chem arxiv}. YEAR = {2023}}
Computing binding affinities is of great importance in drug discovery pipeline and its prediction using advanced machine learning methods still remains a major challenge as the existing datasets and models do not consider the dynamic features of protein-ligand interactions. To this end, we have developed PLAS-20k dataset, an extension of previously developed PLAS-5k, with 97,500 independent simulations on a total of 19,500 different protein-ligand complexes. Our results show good correlation with the available experimental values, performing better than docking scores. This holds true even for a subset of ligands that follows Lipinski’s rule, and for diverse clusters of complex structures, thereby highlighting the importance of PLAS-20k dataset in developing new ML models. Along with this, our dataset is also beneficial in classifying strong and weak binders compared to docking. Further, OnionNet model has been retrained on PLAS-20k dataset and is provided as a baseline for the prediction of binding affinities. We believe that large-scale MD-based datasets along with trajectories will form new synergy, paving the way for accelerating drug discovery.
MolOpt: Autonomous Molecular Geometry Optimization using Multi-Agent Reinforcement Learning
Rohit Modee,Sarvesh Mehta,Siddhartha Laghuvarapu,Deva Priyakumar U
Chem arxiv, Chem arxiv, 2023
@inproceedings{bib_MolO_2023, AUTHOR = {Modee, Rohit and Mehta, Sarvesh and Laghuvarapu, Siddhartha and U, Deva Priyakumar }, TITLE = {MolOpt: Autonomous Molecular Geometry Optimization using Multi-Agent Reinforcement Learning}, BOOKTITLE = {Chem arxiv}. YEAR = {2023}}
In this paper, we propose MolOpt, the first attempt of its kind to use Multi-Agent Reinforcement Learning (MARL) for autonomous molecular geometry optimization (MGO). Typically MGO algorithms are hand-designed, but MolOpt uses MARL to learn a learned optimizer (policy) that can perform MGO without depending on other hand-designed optimizers. We cast MGO as a MARL problem, where each agent corresponds to a single atom in the molecule. MolOpt performs MGO by minimizing the forces on each atom in the molecule. Our experiments demonstrate the generalizing ability of MolOpt for MGO of Propane, Pentane, Heptane, Hexane, and Octane when trained on Ethane, Butane, and Isobutane. In terms of performance, MolOpt outperforms the MDMin optimizer and demonstrates similar performance to the FIRE optimizer. However, it does not surpass the BFGS optimizer. The results demonstrate that MolOpt has the potential to introduce innovative advancements in MGO by providing a novel approach using reinforcement learning (RL), which may open up new research directions for MGO. Overall, this work serves as a proof-of-concept for the potential of MARL in MGO.
MeGen-Generation of gallium metal clusters using Reinforcement Learning
Rohit Modee,Ashwini Verm,Kavita Joshi,Deva Priyakumar U
Machine Learning: Science and Technology, MLST, 2023
@inproceedings{bib_MeGe_2023, AUTHOR = {Modee, Rohit and Verm, Ashwini and Joshi, Kavita and U, Deva Priyakumar }, TITLE = {MeGen-Generation of gallium metal clusters using Reinforcement Learning}, BOOKTITLE = {Machine Learning: Science and Technology}. YEAR = {2023}}
The generation of low-energy 3D structures of metal clusters depends on the efficiency of the search algorithm and the accuracy of inter-atomic interaction description. In this work, we formulate the search algorithm as a Reinforcement Learning (RL) problem. Concisely, we propose a novel actor-critic architecture that generates low-lying isomers of metal clusters at a fraction of computational cost than conventional methods. Our RL-based search algorithm uses a previously developed DART model as a reward function to describe the inter-atomic interactions to validate predicted structures. Using the DART model as a reward function incentivizes the RL model to generate low-energy structures and helps generate valid structures. We demonstrate the advantages of our approach over conventional methods for scanning local minima on potential energy surface (PES). Our approach not only generates isomer of
Single-Lead to Multi-Lead Electrocardiogram Reconstruction Using a Modified Attention U-Net Framework
Akshit Garg,Vijay Vignesh Venkatramani,Deva Priyakumar U
International Joint Conference on Neural Networks, IJCNN, 2023
@inproceedings{bib_Sing_2023, AUTHOR = {Garg, Akshit and Venkatramani, Vijay Vignesh and U, Deva Priyakumar }, TITLE = {Single-Lead to Multi-Lead Electrocardiogram Reconstruction Using a Modified Attention U-Net Framework}, BOOKTITLE = {International Joint Conference on Neural Networks}. YEAR = {2023}}
—With Cardiovascular Diseases on the rise around the world, Electrocardiograms (ECGs) play a crucial role in their diagnosis owing to their non-invasive nature and simplicity. Medical professionals typically use 12-lead ECGs for medical analysis but gathering 12-lead ECG data is an arduous task outside clinical setting. Modern wearables can collect an ECG with fewer leads than the standard 12 leads. However, medical professionals and conventional ECG analysis software find this reduced lead set data challenging to interpret. By using the reduced lead set data to create standard 12-lead ECG data, ECG reconstruction can solve this issue. This paper proposes a novel single-lead to multi-lead ECG reconstruction solution using a modified Attention U-net framework. Using only the lead II of ECG, our model is able to reproduce the other 11 leads of conventional 12-lead ECG with a Pearson correlation, Mean square error and R-squared error of 0.805, 0.0122 and 0.639, respectively. Further, a single combined model is used to reconstruct all 11 leads simultaneously, improving performance and simultaneously reducing the computational resources needed for training compared to current literature in the field. In comparison to previous works, which only reconstruct small ECG segments, our model is trained to reconstruct longer 10-second ECG signals. We demonstrate our model’s ability for real-life utilisation using a cardiovascular disease classification task. A deep learning model was trained for multi-disease classification on actual 12-lead ECG data and was tested on both original and reconstructed 12-lead ECG signals. The classification accuracies for the original and reconstructed signals were comparable, portraying that our reconstruction model can preserve diagnostically relevant artefacts in its reconstructed signals. This work provides a new promising solution in the field of single-lead ECG reconstruction, taking us a step closer to bridging the divide between reduced lead set data and existing 12-lead ECG end users like clinicians and automatic ECG classifiers. Index Terms—Healthcare, Cardiology, Electrocardiograms, Machine Learning, Neural Networks
PREHOST: Host prediction of coronaviridae family using machine learning
Anusha Chaturvedi,Kushal Borkar,Deva Priyakumar U,Vinod Palakkad Krishnanunni
Heliyon, HEL, 2023
@inproceedings{bib_PREH_2023, AUTHOR = {Chaturvedi, Anusha and Borkar, Kushal and U, Deva Priyakumar and Krishnanunni, Vinod Palakkad }, TITLE = {PREHOST: Host prediction of coronaviridae family using machine learning}, BOOKTITLE = {Heliyon}. YEAR = {2023}}
Coronavirus, a zoonotic virus capable of transmitting infections from animals to humans, emerged as a pandemic recently. In such circumstances, it is essential to understand the virus’s origin. In this study, we present a novel machine-learning pipeline PreHost for host prediction of the family, Coronaviridae. We leverage the complete viral genome and sequences at the protein level (spike protein, membrane protein, and nucleocapsid protein). Compared with the current state-of-the-art approaches, the random forest model attained high accuracy and recall scores of 99.91% and 0.98, respectively, for genome sequences. In addition to the spike protein sequences, our study shows membrane and nucleocapsid protein sequences can be utilized to predict the host of viruses. We also identified important sites in the viral sequences that help distinguish between different host classes. The host prediction pipeline PreHost will cater as a valuable tool to take effective measures to govern the transmission of future viruses.
Carbodicarbenes and Striking Redox Transitions of their Conjugate Acids: Influence of NHC versus CAAC as Donor Substituents
Ramapada Dolai,Rahul Kumar,Benedict J Elvers,Pradeep Kumar Pal,Benson Joseph,Rina Sikari,Mithilesh Kumar Nayak,Avijit Maiti,Deva Priyakumar U
Chemistry–A European Journal, CAEJ, 2023
@inproceedings{bib_Carb_2023, AUTHOR = {Dolai, Ramapada and Kumar, Rahul and Elvers, Benedict J and Pal, Pradeep Kumar and Joseph, Benson and Sikari, Rina and Nayak, Mithilesh Kumar and Maiti, Avijit and U, Deva Priyakumar }, TITLE = {Carbodicarbenes and Striking Redox Transitions of their Conjugate Acids: Influence of NHC versus CAAC as Donor Substituents}, BOOKTITLE = {Chemistry–A European Journal}. YEAR = {2023}}
Herein, a new type of carbodicarbene (CDC) comprising two different classes of carbenes is reported; NHC and CAAC as donor substituents and compare the molecular structure and coordination to Au(I)Cl to those of NHC-only and CAAC-only analogues. The conjugate acids of these three CDCs exhibit notable redox properties. Their reactions with [NO][SbF6] were investigated. The reduction of the conjugate acid of CAAC-only based CDC with KC8 results in the formation of hydrogen abstracted/eliminated products, which proceed through a neutral radical intermediate, detected by EPR spectroscopy. In contrast, the reduction of conjugate acids of NHC-only and NHC/CAAC based CDCs led to intermolecular reductive (reversible) carbon–carbon sigma bond formation. The resulting relatively elongated carbon– carbon sigma bonds were found to be readily oxidized. They were, thus, demonstrated to be potent reducing agents, underlining their potential utility as organic electron donors and n-dopants in organic semiconductor molecules.
Efficient and enhanced sampling of drug‐like chemical space for virtual screening and molecular design using modern machine learning methods
Manan Goel,Rishal Aggarwal,S Bhuvanesh,Pradeep Kumar Pal,Deva Priyakumar U
Wiley Interdisciplinary Reviews: Computational Molecular Science, WCMS, 2023
Abs | | bib Tex
@inproceedings{bib_Effi_2023, AUTHOR = {Goel, Manan and Aggarwal, Rishal and Bhuvanesh, S and Pal, Pradeep Kumar and U, Deva Priyakumar }, TITLE = {Efficient and enhanced sampling of drug‐like chemical space for virtual screening and molecular design using modern machine learning methods}, BOOKTITLE = {Wiley Interdisciplinary Reviews: Computational Molecular Science}. YEAR = {2023}}
Drug design involves the process of identifying and designing novel molecules that have desirable properties and bind well to a given target receptor. Typically, such molecules are identified by screening large chemical libraries for desirable physicochemical properties and binding strength with the target protein. This traditional approach, however, has severe limitations as exhaustively screening every molecule in known chemical libraries is computationally infeasible. Furthermore, currently available molecular libraries are only a minuscule part of the entire set of possible drug‐like molecular structures (drug‐like chemical space). In this review, we discuss how the former limitation is addressed by modeling virtual screening as a search space problem and how these endeavors utilize machine learning to reduce the number of required computational experiments to identify top candidates. We follow that up by …
Modified Variable Kernel Length ResNets for Heart Murmur Detection and Clinical Outcome Prediction using Multi-positional Phonocardiogram Recording
V Vijay Vignesh,Akshit Garg,Deva Priyakumar U
Computing in Cardiology, CinC, 2022
@inproceedings{bib_Modi_2022, AUTHOR = {Vignesh, V Vijay and Garg, Akshit and U, Deva Priyakumar }, TITLE = {Modified Variable Kernel Length ResNets for Heart Murmur Detection and Clinical Outcome Prediction using Multi-positional Phonocardiogram Recording}, BOOKTITLE = {Computing in Cardiology}. YEAR = {2022}}
In this work, we describe an end-to-end deep learning
architecture for Heart Murmur Detection from Phonocardiogram(PCG) recordings as part of The George B. Moody
PhysioNet Challenge 2022. Our team, “Team IIITH” received a weighted accuracy score of 0.708 (ranked 19th out
of 40 teams) and Challenge cost of 13264 (ranked 22nd out
of 39 teams) on the official hidden test set.
In our approach, the PCG recordings are first downsampled to 1000 Hz before being passed through a Butterworth’s low and high pass filter to remove baseline
wanders and high-frequency noise present in the recordings. The PCG recordings are then broken down into
10-second segments and normalized to bring all trainable samples to the same size. To extract embeddings
more efficiently, we built a custom 1-dimensional Residual
Network (ResNet) where the 10-second inputs are passed
through variable-sized kernel ResNets in parallel, before
being concatenated and passed through the next ResNet
layer to account for different length dependencies across
the PCG signal. The output of this custom ResNet is then
fed to a 2-layer feed-forward network for final classification. Cross-Entropy Loss with class weights was employed
to account for class imbalance. Our approach obtained
a 5-fold Cross-Validation weighted accuracy score of 0.71
and challenge cost score of 12067 on the training set
Modified Variable Kernel Length ResNets for Heart Murmur Detection and Clinical Outcome Prediction Using Phonocardiogram Recordings
Vijay Vignesh Venkataramani,Akshit Garg,Deva Priyakumar U
Computing in Cardiology, CinC, 2022
@inproceedings{bib_Modi_2022, AUTHOR = {Venkataramani, Vijay Vignesh and Garg, Akshit and U, Deva Priyakumar }, TITLE = {Modified Variable Kernel Length ResNets for Heart Murmur Detection and Clinical Outcome Prediction Using Phonocardiogram Recordings}, BOOKTITLE = {Computing in Cardiology}. YEAR = {2022}}
In this work, we describe an end-to-end deep learning architecture for Heart Murmur Detection from Phonocardiogram(PCG) recordings as part of The George B. Moody PhysioNet Challenge 2022. Our team, “Team IIITH” received a weighted accuracy score of 0.708 (ranked 19th out of 40 teams) and Challenge cost of 13264 (ranked 22nd out of 39 teams) on the official hidden test set. In our approach, the PCG recordings are first downsampled to 1000 Hz before being passed through a Butterworth’s low and high pass filter to remove baseline wanders and high-frequency noise present in the recordings. The PCG recordings are then broken down into 10-second segments and normalized to bring all trainable samples to the same size. To extract embeddings more efficiently, we built a custom 1-dimensional Residual Network (ResNet) where the 10-second inputs are passed through variable-sized kernel ResNets in parallel, before being concatenated and passed through the next ResNet layer to account for different length dependencies across the PCG signal. The output of this custom ResNet is then fed to a 2-layer feed-forward network for final classification. Cross-Entropy Loss with class weights was employed to account for class imbalance. Our approach obtained a 5-fold Cross-Validation weighted accuracy score of 0.71 and challenge cost score of 12067 on the training set.
Staufen-2 functions as a cofactor for enhanced Rev-mediated nucleocytoplasmic trafficking of HIV-1 genomic RNA via the CRM1 pathway
Kannan Balakrishnan, Punnagai Munusami,Krishnaveni Mohareer,Deva Priyakumar U, Atoshi Banerjee,Tom Luedde,Shekhar C. Mande, Carsten Münk, Sharmistha Banerjee
FEBS Journal, FEBSJ, 2022
Abs | | bib Tex
@inproceedings{bib_Stau_2022, AUTHOR = {Balakrishnan, Kannan and Munusami, Punnagai and Mohareer, Krishnaveni and U, Deva Priyakumar and Banerjee, Atoshi and Luedde, Tom and Mande, Shekhar C. and Münk, Carsten and Banerjee, Sharmistha }, TITLE = {Staufen-2 functions as a cofactor for enhanced Rev-mediated nucleocytoplasmic trafficking of HIV-1 genomic RNA via the CRM1 pathway}, BOOKTITLE = {FEBS Journal}. YEAR = {2022}}
Nucleocytoplasmic shuttling of viral elements, supported by several host factors, is essential for the replication of the human immunodeficiency virus (HIV). HIV-1 uses a nuclear RNA export pathway mediated by viral protein Rev to transport its Rev response element (RRE)-containing partially spliced and unspliced transcripts aided by the host nuclear RNA export protein CRM1. The factor(s) interacting with the CRM1-Rev complex are potential antiretroviral target(s) and could serve as a retroviral model system to study nuclear export machinery adapted by these viruses. We earlier reported that cellular Staufen-2 interacts with Rev, facilitating viral-RNA export. Here, we identified the formation of a complex between Staufen-2, CRM1 and Rev. Molecular docking and simulations mapped the interacting residues in the RNA-binding Domain 4 of Staufen-2 as R336 and R337, which were experimentally verified to be critical for interactions among Staufen-2, CRM1 and Rev by mutational analysis. Staufen-2 mutants defective in interaction with CRM1 or Rev failed to supplement the Rev-RNA export activity and viral production, demonstrating the importance of these interactions. Rev-dependent reporter assays and proviral DNA-construct transfection-based studies in Staufen-2 knockout cells in the presence of leptomycin-B (LMB) revealed a significant reduction in CRM1-mediated Rev-dependent RNA export with decreased virus production as compared to Staufen-2 knockout background or LMB treatment alone, suggesting the relevance of these interactions in augmenting RNA export activity of Rev. Our observations provide further insights into the mechanistic intricacies of unspliced viral-RNA export to the cytoplasm and support the notion that abrogating such interactions can reduce HIV-1 proliferation.
Mining subgraph coverage patterns from graph transactions
Srinivas Reddy Annappalli,Krishna Reddy Polepalli,Anirban Mondal,Deva Priyakumar U
International Journal of Data Science and Analytics, IJDSA, 2022
Abs | | bib Tex
@inproceedings{bib_Mini_2022, AUTHOR = {Annappalli, Srinivas Reddy and Polepalli, Krishna Reddy and Mondal, Anirban and U, Deva Priyakumar }, TITLE = {Mining subgraph coverage patterns from graph transactions}, BOOKTITLE = {International Journal of Data Science and Analytics}. YEAR = {2022}}
Pattern mining from graph transactional data (GTD) is an active area of research with applications in the domains of bioinformatics, chemical informatics and social networks. Existing works address the problem of mining frequent subgraphs from GTD. However, the knowledge concerning the coverage aspect of a set of subgraphs is also valuable for improving the performance of several applications. In this regard, we introduce the notion of subgraph coverage patterns (SCPs). Given a GTD, a subgraph coverage pattern is a set of subgraphs subject to relative frequency, coverage and overlap constraints provided by the user. We propose the Subgraph ID-based Flat Transactional (SIFT) framework for the efficient extraction of SCPs from a given GTD. Our performance evaluation using three real datasets demonstrates that our proposed SIFT framework is indeed capable of efficiently extracting SCPs from GTD. Furthermore, we demonstrate the effectiveness of SIFT through a case study in computer-aided drug design.
MolGPT: Molecular Generation Using a Transformer-Decoder Model
Viraj Bagal,Rishal Aggarwal,Vinod Palakkad Krishnanunni,Deva Priyakumar U
Journal of Chemical Information and Modeling, JCIM, 2022
@inproceedings{bib_MolG_2022, AUTHOR = {Bagal, Viraj and Aggarwal, Rishal and Krishnanunni, Vinod Palakkad and U, Deva Priyakumar }, TITLE = {MolGPT: Molecular Generation Using a Transformer-Decoder Model}, BOOKTITLE = {Journal of Chemical Information and Modeling}. YEAR = {2022}}
Application of deep learning techniques for de novo generation of molecules, termed as inverse molecular design, has been gaining enormous traction in drug design. The representation of molecules in SMILES notation as a string of characters enables the usage of state of the art models in natural language processing, such as Transformers, for molecular design in general. Inspired by generative pre-training (GPT) models that have been shown to be successful in generating meaningful text, we train a transformer-decoder on the next token prediction task using masked self-attention for the generation of druglike molecules in this study. We show that our model, MolGPT, performs on par with other previously proposed modern machine learning frameworks for molecular generation in terms of generating valid, unique, and novel molecules. Furthermore, we demonstrate that the model can be trained conditionally to
Modern machine learning for tackling inverse problems in chemistry: molecular design to realization
S Bhuvanesh,Manan Goel,Deva Priyakumar U
Chemical communications, C C, 2022
Abs | | bib Tex
@inproceedings{bib_Mode_2022, AUTHOR = {Bhuvanesh, S and Goel, Manan and U, Deva Priyakumar }, TITLE = {Modern machine learning for tackling inverse problems in chemistry: molecular design to realization}, BOOKTITLE = {Chemical communications}. YEAR = {2022}}
The discovery of new molecules and materials helps expand the horizons of novel and innovative real-life applications. In pursuit of finding molecules with desired properties, chemists have traditionally relied on experimentation and recently on combinatorial methods to generate new substances often complimented by computational methods. The sheer size of the chemical space makes it infeasible to search through all possible molecules exhaustively. This calls for fast and efficient methods to navigate the chemical space to find substances with desired properties. This class of problems is referred to as inverse design problems. There are a variety of inverse problems in chemistry encompassing various subfields like drug discovery, retrosynthesis, structure identification, etc. Recent developments in modern machine learning (ML) methods have shown great promise in tackling problems of this kind. This has …
MO-MEMES: A method for accelerating virtual screening using multi-objective Bayesian optimization
Sarvesh Mehta,Manan Goel,Deva Priyakumar U
Frontiers in Medicine, FIM, 2022
Abs | | bib Tex
@inproceedings{bib_MO-M_2022, AUTHOR = {Mehta, Sarvesh and Goel, Manan and U, Deva Priyakumar }, TITLE = {MO-MEMES: A method for accelerating virtual screening using multi-objective Bayesian optimization}, BOOKTITLE = {Frontiers in Medicine}. YEAR = {2022}}
Drug discovery is a long, expensive, and extremely laborious process that involves multiple steps with knowledge from a wide variety of domains like chemistry, biology and pharmacology. The first step in this process is the identification of potential hit molecules for a novel target followed by experimental evaluation typically using biochemical assays toward lead identification. These hits are then optimized to have higher binding affinity, low toxicity, and improved bioavailability among other requirements. The time and expense involved in this process has given rise to alternate in silico approaches like virtual screening wherein molecules are computationally evaluated to identify potential hits. The structure based drug design (SBDD) method, docking, is used most commonly in virtual screening to identify molecules with high binding affinity to the given target (1–
Modern AI/ML Methods for Healthcare: Opportunities and Challenges
Akshit Garg,Vijay Vignesh Venkataramani,Akshaya Karthikeyan,Deva Priyakumar U
International Conference on Distributed Computing and Internet Technology, ICDCIT, 2022
Abs | | bib Tex
@inproceedings{bib_Mode_2022, AUTHOR = {Garg, Akshit and Venkataramani, Vijay Vignesh and Karthikeyan, Akshaya and U, Deva Priyakumar }, TITLE = {Modern AI/ML Methods for Healthcare: Opportunities and Challenges}, BOOKTITLE = {International Conference on Distributed Computing and Internet Technology}. YEAR = {2022}}
Artificial Intelligence has seen a significant resurgence in the past decade in wide ranging technology and domain areas. Recent progress in digitisation and high influx of biomedical data have led to an unparalleled success of Machine Learning systems in healthcare, which is perceived to be a possible game changer for ‘healthcare to all’. This article gives an account of some of the current applications of AI solutions in the medical domains of diagnosis, prognosis and treatment. The article will also illustrate the implications of AI in the fight against the COVID-19 pandemic. Lastly, the article will summarise the challenges AI currently faces in its wide-scale adoption in the healthcare industry and how they can possibly be dealt with to move towards a more intelligent medical future. This may enable moving towards quality healthcare for all.
Benchmark study on deep neural network potentials for small organic molecules
Rohit Modee,Siddhartha Laghuvarapu,Deva Priyakumar U
Journal of Computational Chemistry, JCC, 2022
Abs | | bib Tex
@inproceedings{bib_Benc_2022, AUTHOR = {Modee, Rohit and Laghuvarapu, Siddhartha and U, Deva Priyakumar }, TITLE = {Benchmark study on deep neural network potentials for small organic molecules}, BOOKTITLE = {Journal of Computational Chemistry}. YEAR = {2022}}
There has been tremendous advancement in machine learning (ML) applications in computational chemistry, particularly in neural network potentials (NNP). NNPs can approximate potential energy surface (PES) as a high dimensional function by learning from existing reference data, thereby circumventing the need to solve the electronic Schrödinger equation explicitly. As a result, ML accelerates chemical space exploration and property prediction compared to quantum mechanical methods. Novel ML methods have the potential to provide efficient means for predicting the properties of molecules. However, this potential has been limited by the lack of standard comparative evaluations. In this work, we compare four selected models, that is, ANI, PhysNet, SchNet, and BAND‐NN, developed to represent the PES of small organic molecules. We evaluate these models for their accuracy and transferability on two
BiRDS-Binding Residue Detection from Protein Sequences using Deep ResNets
Vineeth Ravindra Chelur,Deva Priyakumar U
Journal of Chemical Information and Modeling, JCIM, 2022
@inproceedings{bib_BiRD_2022, AUTHOR = {Chelur, Vineeth Ravindra and U, Deva Priyakumar }, TITLE = {BiRDS-Binding Residue Detection from Protein Sequences using Deep ResNets}, BOOKTITLE = {Journal of Chemical Information and Modeling}. YEAR = {2022}}
Protein–drug interactions play important roles in many biological processes and therapeutics. Predicting the binding sites of a protein helps to discover such interactions. New drugs can be designed to optimize these interactions, improving protein function. The tertiary structure of a protein decides the binding sites available to the drug molecule, but the determination of the 3D structure is slow and expensive. Conversely, the determination of the amino acid sequence is swift and economical. Although quick and accurate prediction of the binding site using just the sequence is challenging, the application of Deep Learning, which has been hugely successful in several biochemical tasks, makes it feasible. BiRDS is a Residual Neural Network that predicts the protein’s most active binding site using sequence information. SC-PDB, an annotated database of druggable binding sites, is used for training the network …
System and method for exploring chemical space during molecular design using a machine learning model
Deva Priyakumar U,Sarvesh Mehta,Siddhartha Laghuvarapu,Yashaswi Pathak
United States Patent, Us patent, 2022
@inproceedings{bib_Syst_2022, AUTHOR = {U, Deva Priyakumar and Mehta, Sarvesh and Laghuvarapu, Siddhartha and Pathak, Yashaswi }, TITLE = {System and method for exploring chemical space during molecular design using a machine learning model}, BOOKTITLE = {United States Patent}. YEAR = {2022}}
A system and method for exploring a chemical space during molecular design for at least one top hit molecule using a machine learning (ML) model are provided. The method includes (i) representing the at least one molecule stored in a drug library into at least one vector;(ii) clustering the at least one vector to obtain at least one cluster of molecules into one or more clusters;(iii) uniformly sampling a first subset of molecules from each cluster of molecules;(vi) determining a docking score for sampled subset of molecules;(iv) training the ML model by correlating sampled subset of molecules with docking score;(viii) computing acquisition function values for a second subset of molecules from each cluster; and (ix) determining at least one top hit molecule based on the computed acquisition function values, thereby exploring the chemical space for the at least one top hit molecule.
Deep Reinforcement Learning for Molecular Inverse Problem of Nuclear Magnetic Resonance Spectra to Molecular Structure
S Bhuvanesh,Sarvesh Mehta,Yashaswi Pathak,Deva Priyakumar U
The Journal of Physical Chemistry Letters, JPCL, 2022
Abs | | bib Tex
@inproceedings{bib_Deep_2022, AUTHOR = {Bhuvanesh, S and Mehta, Sarvesh and Pathak, Yashaswi and U, Deva Priyakumar }, TITLE = {Deep Reinforcement Learning for Molecular Inverse Problem of Nuclear Magnetic Resonance Spectra to Molecular Structure}, BOOKTITLE = {The Journal of Physical Chemistry Letters}. YEAR = {2022}}
Spectroscopy is the study of how matter interacts with electromagnetic radiation. The spectra of any molecule are highly information-rich, yet the inverse relation of spectra to the corresponding molecular structure is still an unsolved problem. Nuclear magnetic resonance (NMR) spectroscopy is one such critical technique in the scientists’ toolkit to characterize molecules. In this work, a novel machine learning framework is proposed that attempts to solve this inverse problem by navigating the chemical space to find the correct structure given an NMR spectra. The proposed framework uses a combination of online Monte Carlo tree search (MCTS) and a set of graph convolution networks to build a molecule iteratively. Our method can predict the structure of the molecule ∼80% of the time in its top 3 guesses for molecules with <10 heavy atoms. We believe that the proposed framework is a significant step in solving the
Synthesis of α‐Aryl Ketones by Harnessing the Non‐Innocence of Toluene and its Derivatives: Enhancing the Acidity of Methyl Arenes by a Brønsted Base and their Mechanistic Aspects
Ramdas Sreedharan,Pradeep Kumar Pal,Pradeep Kumar Reddy Panyam,Deva Priyakumar U,Thirumanavelan Gandhi
Asian Journal of Organic Chemistry, AJOC, 2022
Abs | | bib Tex
@inproceedings{bib_Synt_2022, AUTHOR = {Sreedharan, Ramdas and Pal, Pradeep Kumar and Panyam, Pradeep Kumar Reddy and U, Deva Priyakumar and Gandhi, Thirumanavelan }, TITLE = {Synthesis of α‐Aryl Ketones by Harnessing the Non‐Innocence of Toluene and its Derivatives: Enhancing the Acidity of Methyl Arenes by a Brønsted Base and their Mechanistic Aspects}, BOOKTITLE = {Asian Journal of Organic Chemistry}. YEAR = {2022}}
Ketones are the key functional group that recurs in chemistry and biology, and accessing them through simple and economic ways is highly desirable. Herein, we report the synthesis of unsymmetrical ketones from abundant toluene and alkyl esters, where volatile alcohols are the sole byproduct. This protocol applies to a repertoire of substrates bearing electron‐donating, electron‐withdrawing, and neutral substituents. Most importantly, the organometallic ferrocenyl ester underwent aroylation with ease. This method is the first example to furnish diketones from methyl arenes and diesters. Furthermore, cyclic imide was synthesized by this protocol utilizing KN(SiMe3)2 as a ′nitrogen′ source. Density functional theory studies provide insight into deprotonation of toluene by K+‐π interaction by increasing its acidity, and this being the rate‐determining step.
Plas-5k: Dataset of protein-ligand affinities from molecular dynamics for machine learning applications
Divya B Kolepara,Charuvaka Muvva,S Bhuvanesh,Akshit Garg,Rohit Modee,Agastya P. Bhati,Divya Nayar,C. S.Vasavi,Shruti S Jeurkar,Pradeep Kumar Pal,Subhajit Roy,Sarvesh Mehta,Shubham Sharma,Vishal Kumar,Deva Priyakumar U
Scientific Data, SD, 2022
@inproceedings{bib_Plas_2022, AUTHOR = {Kolepara, Divya B and Muvva, Charuvaka and Bhuvanesh, S and Garg, Akshit and Modee, Rohit and Bhati, Agastya P. and Nayar, Divya and S.Vasavi, C. and Jeurkar, Shruti S and Pal, Pradeep Kumar and Roy, Subhajit and Mehta, Sarvesh and Sharma, Shubham and Kumar, Vishal and U, Deva Priyakumar }, TITLE = {Plas-5k: Dataset of protein-ligand affinities from molecular dynamics for machine learning applications}, BOOKTITLE = {Scientific Data}. YEAR = {2022}}
Computational methods and recently modern machine learning methods have played a key role in structure-based drug design. Though several benchmarking datasets are available for machine learning applications in virtual screening, accurate prediction of binding afnity for a protein-ligand complex remains a major challenge. New datasets that allow for the development of models for predicting binding afnities better than the state-of-the-art scoring functions are important. For the frst time, we have developed a dataset, PLAS-5k comprised of 5000 protein-ligand complexes chosen from PDB database. The dataset consists of binding afnities along with energy components like electrostatic, van der Waals, polar and non-polar solvation energy calculated from molecular dynamics simulations using MMPBSA (Molecular Mechanics Poisson-Boltzmann Surface Area) method. The calculated binding afnities outperformed docking scores and showed a good correlation with the available experimental values. The availability of energy components may enable optimization of desired components during machine learning-based drug design. Further, OnionNet model has been retrained on PLAS-5k dataset and is provided as a baseline for the prediction of binding afnities
Tetra-Coordinated Boron-Functionalized Phenanthroimidazole-Based Zinc Salen as a Photocatalyst for the Cycloaddition of CO2 and Epoxides
Prakash Nayak,Anna Chandrasekar Murali,Pradeep Kumar Pal,Deva Priyakumar U,chandra sekhar vadapalli,Krishnan Venkatasubbaiah
Inorganic Chemistry, ICh, 2022
Abs | | bib Tex
@inproceedings{bib_Tetr_2022, AUTHOR = {Nayak, Prakash and Murali, Anna Chandrasekar and Pal, Pradeep Kumar and U, Deva Priyakumar and Vadapalli, chandra Sekhar and Venkatasubbaiah, Krishnan }, TITLE = {Tetra-Coordinated Boron-Functionalized Phenanthroimidazole-Based Zinc Salen as a Photocatalyst for the Cycloaddition of CO2 and Epoxides}, BOOKTITLE = {Inorganic Chemistry}. YEAR = {2022}}
A unique B–N coordinated phenanthroimidazole-based zinc salen was synthesized. The zinc salen thus synthesized acts as a photocatalyst for the cycloaddition of carbon dioxide with terminal epoxides under ambient conditions. DFT study of the cycloaddition of carbon dioxide with terminal epoxide indicates the preference of the reaction pathway when photocatalyzed by zinc salen. We anticipate that this strategy will help to design new photocatalysts for CO2 fixation.
Staufen‐2 functions as a cofactor for enhanced Rev‐mediated nucleocytoplasmic trafficking of HIV‐1 genomic RNA via the CRM1 pathway
Kannan Balakrishnan,Punnagai Munusami,Krishnaveni Mohareer,Deva Priyakumar U,Atoshi Banerjee,Tom Luedde,Shekhar C Mande,Carsten Münk,Sharmistha Banerjee,Sharmistha Banerjee
FEBS Journal, FEBSJ, 2022
Abs | | bib Tex
@inproceedings{bib_Stau_2022, AUTHOR = {Balakrishnan, Kannan and Munusami, Punnagai and Mohareer, Krishnaveni and U, Deva Priyakumar and Banerjee, Atoshi and Luedde, Tom and Mande, Shekhar C and Münk, Carsten and Banerjee, Sharmistha and Banerjee, Sharmistha }, TITLE = {Staufen‐2 functions as a cofactor for enhanced Rev‐mediated nucleocytoplasmic trafficking of HIV‐1 genomic RNA via the CRM1 pathway}, BOOKTITLE = {FEBS Journal}. YEAR = {2022}}
Nucleocytoplasmic shuttling of viral elements, supported by several host factors, is essential for the replication of the human immunodeficiency virus (HIV). HIV‐1 uses a nuclear RNA export pathway mediated by viral protein Rev to transport its Rev response element (RRE)‐containing partially spliced and unspliced transcripts aided by the host nuclear RNA export protein CRM1. The factor(s) interacting with the CRM1‐Rev complex are potential antiretroviral target(s) and could serve as a retroviral model system to study nuclear export machinery adapted by these viruses. We earlier reported that cellular Staufen‐2 interacts with Rev, facilitating viral‐RNA export. Here, we identified the formation of a complex between Staufen‐2, CRM1 and Rev. Molecular docking and simulations mapped the interacting residues in the RNA‐binding Domain 4 of Staufen‐2 as R336 and R337, which were experimentally verified to be …
COVID-19 Risk Stratification and Mortality Prediction in Hospitalized Indian Patients: Harnessing clinical data for public health benefits
Shanmukh Alle,Akshay Kanakan,Samreen Siddiqui,Akshit Garg,Akshaya Karthikeyan,Priyanka Mehta,Neha Mishra,Vinod Palakkad Krishnanunni,Deva Priyakumar U
Plos One, Plos One, 2022
@inproceedings{bib_COVI_2022, AUTHOR = {Alle, Shanmukh and Kanakan, Akshay and Siddiqui, Samreen and Garg, Akshit and Karthikeyan, Akshaya and Mehta, Priyanka and Mishra, Neha and Krishnanunni, Vinod Palakkad and U, Deva Priyakumar }, TITLE = {COVID-19 Risk Stratification and Mortality Prediction in Hospitalized Indian Patients: Harnessing clinical data for public health benefits}, BOOKTITLE = {Plos One}. YEAR = {2022}}
The variability of clinical course and prognosis of COVID-19 highlights the necessity of patient sub-group risk stratification based on clinical data. In this study, clinical data from a cohort of Indian COVID-19 hospitalized patients is used to develop risk stratification and mortality prediction models. We analyzed a set of 70 clinical parameters including physiological and hematological for developing machine learning models to identify biomarkers. We also compared the Indian and Wuhan cohort, and analyzed the role of steroids. A bootstrap averaged ensemble of Bayesian networks was also learned to construct an explainable model for discovering actionable influences on mortality and days to outcome. We discovered blood parameters, diabetes, co-morbidity and SpO2 levels as important risk stratification features, whereas mortality prediction is dependent only on blood parameters. XGboost and logistic regression model yielded the best performance on risk stratification and mortality prediction, respectively (AUC score 0.83, AUC score 0.92). Blood coagulation parameters (ferritin, D-Dimer and INR), immune and inflammation parameters IL6, LDH and Neutrophil (%) are common features for both risk and mortality prediction. Compared with Wuhan patients, Indian patients with extreme blood parameters indicated higher survival rate. Analyses of medications suggest that a higher proportion of survivors and mild patients who were administered steroids had extreme neutrophil and lymphocyte percentages. The ensemble averaged Bayesian network structure revealed serum ferritin to be the mos
Structure-based drug repurposing: Traditional and advanced AI/ML-aided methods
Chinmayee Choudhury,N. Arul Murugan,Deva Priyakumar U
Drug Discovery Today, DDT, 2022
@inproceedings{bib_Stru_2022, AUTHOR = {Choudhury, Chinmayee and Murugan, N. Arul and U, Deva Priyakumar }, TITLE = {Structure-based drug repurposing: Traditional and advanced AI/ML-aided methods}, BOOKTITLE = {Drug Discovery Today}. YEAR = {2022}}
The current global health emergency in the form of the Coronavirus 2019 (COVID-19) pandemic has 16 highlighted the need for fast, accurate, and efficient drug discovery pipelines. Traditional drug 17 discovery projects relying on in vitro high-throughput screening (HTS) involve large investments and 18 sophisticated experimental set-ups, affordable only to big biopharmaceutical companies. In this 19 scenario, application of efficient state-of-the-art computational methods and modern artificial 20 intelligence (AI)-based algorithms for rapid screening of repurposable chemical space [approved drugs 21 and natural products (NPs) with proven pharmacokinetic profiles] to identify the initial leads is a 22 powerful option to save resources and time. Structure-based drug repurposing is a popular in silico 23 repurposing approach. In this review, we discuss traditional and modern AI-based computational 24 methods and tools applied at various stages for structure-based drug discovery (SBDD) pipelines. 25 Additionally, we highlight the role of generative m
Synthesis and reactivity of NHC-coordinated phosphinidene oxide
Debabrata Dhara , Vadapalli Chandrasekhar,Deva Priyakumar U, Anukul Jana,Pradeep Kumar Pal,Ramapada Dolai , Nicolas Chrysochos, Hemant Rawat, Benedict J. Elvers, Ivo Krummenacher, Holger Braunschweig, Carola Schulzke
Chemical communications, C C, 2021
Abs | | bib Tex
@inproceedings{bib_Synt_2021, AUTHOR = {, Debabrata Dhara and Chandrasekhar, Vadapalli and U, Deva Priyakumar and Jana, Anukul and Pal, Pradeep Kumar and , Ramapada Dolai and Chrysochos, Nicolas and Rawat, Hemant and Elvers, Benedict J. and Krummenacher, Ivo and Braunschweig, Holger and Schulzke, Carola }, TITLE = {Synthesis and reactivity of NHC-coordinated phosphinidene oxide}, BOOKTITLE = {Chemical communications}. YEAR = {2021}}
Here we report the synthesis of an N-heterocyclic carbene (NHC)-stabilised phosphinidene oxide by the controlled oxygenation of a phosphinidene under ambient conditions. This compound can be further oxygenated to a phosphinidene dioxide. The stoichiometric reduction of a phosphinidene oxide with KC8 resembles the pinacol coupling reaction–the reduction of a carbonyl compound. We also looked at the stoichiometric oxidation of NHC-coordinated phosphinidene, phosphinidene oxide and phosphinidene dioxide with [NO][SbF6].
SCONES: Self-Consistent Neural Network for Protein Stability Prediction upon Mutation
Yashas Samaga B L,Shampa Raghunathan,Deva Priyakumar U
The Journal of Physical Chemistry B, JPCB, 2021
Abs | | bib Tex
@inproceedings{bib_SCON_2021, AUTHOR = {L, Yashas Samaga B and Raghunathan, Shampa and U, Deva Priyakumar }, TITLE = {SCONES: Self-Consistent Neural Network for Protein Stability Prediction upon Mutation}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2021}}
Engineering proteins to have desired properties by mutating amino acids at specific sites is commonplace. Such engineered proteins must be stable to function. Experimental methods used to determine stability at throughputs required to scan the protein sequence space thoroughly are laborious. To this end, many machine learning based methods have been developed to predict thermodynamic stability changes upon mutation. These methods have been evaluated for symmetric consistency by testing with hypothetical reverse mutations. In this work, we propose transitive data augmentation, evaluating transitive consistency with our new Stransitive data set, and a new machine learning based method, the first of its kind, that incorporates both symmetric and transitive properties into the architecture. Our method, called SCONES, is an interpretable neural network that predicts small relative protein stability changes for missense mutations that do not significantly alter the structure. It estimates a residue’s contributions toward protein stability (ΔG) in its local structural environment, and the difference between independently predicted contributions of the reference and mutant residues is reported as ΔΔG. We show that this self-consistent machine learning architecture is immune to many common biases in data sets, relies less on data than existing methods, is robust to overfitting, and can explain a substantial portion of the variance in experimental data.
DART: Deep learning enabled topological interaction model for energy prediction of metal clusters and its application in identifying unique low energy isomers
Rohit Modee,Sheena Agarwal ,Ashwini Verma ,Kavita Joshi ,Deva Priyakumar U
Physical Chemistry Chemical Physics, PCCP, 2021
Abs | | bib Tex
@inproceedings{bib_DART_2021, AUTHOR = {Modee, Rohit and , Sheena Agarwal and , Ashwini Verma and , Kavita Joshi and U, Deva Priyakumar }, TITLE = {DART: Deep learning enabled topological interaction model for energy prediction of metal clusters and its application in identifying unique low energy isomers}, BOOKTITLE = {Physical Chemistry Chemical Physics}. YEAR = {2021}}
Recently, machine learning (ML) has proven to yield fast and accurate predictions of chemical properties to accelerate the discovery of novel molecules and materials. The majority of the work is on organic molecules, and much more work needs to be done for inorganic molecules, especially clusters. In the present work, we introduce a simple topological atomic descriptor called TAD, which encodes chemical environment information of each atom in the cluster. TAD is a simple and interpretable descriptor where each value represents the atom count in three shells. We also introduce the DART deep learning enabled topological interaction model, which uses TAD as a feature vector to predict energies of metal clusters, in our case gallium clusters with sizes ranging from 31 to 70 atoms. The DART model is designed based on the principle that the energy is a function of atomic interactions and allows us to model these complex atomic interactions to predict the energy. We further introduce a new dataset called GNC_31–70, which comprises structures and DFT optimized energies of gallium clusters with sizes ranging from 31 to 70 atoms. We show how DART can be used to accelerate the process of identification of low energy structures without geometry optimization. Albeit using a topological descriptor, DART achieves a mean absolute error (MAE) of 3.59 kcal mol−1 (0.15 eV) on the test set. We also show that our model can distinguish core and surface atoms in the Ga-70 cluster, which the model has never encountered earlier. Finally, we demonstrate the transferability of the DART model by predicting energies for about 6k unseen configurations picked up from molecular dynamics (MD) data for three cluster sizes (46, 57, and 60) within seconds. The DART model was able to reduce the load on DFT optimizations while identifying unique low energy structures from MD data.
Molecular representations for machine learning applications in chemistry
Shampa Raghunathan,Deva Priyakumar U
International Journal of Quantum Chemistry, IJQC, 2021
@inproceedings{bib_Mole_2021, AUTHOR = {Raghunathan, Shampa and U, Deva Priyakumar }, TITLE = {Molecular representations for machine learning applications in chemistry}, BOOKTITLE = {International Journal of Quantum Chemistry}. YEAR = {2021}}
Machine learning (ML) methods enable computers to address problems by learning from existing data. Such applications are becoming commonplace in molecular sciences. Interest in applying ML techniques across chemical compound space, from predicting properties to designing molecules and materials is in the surge. Especially, ML models have started to accelerate computational chemistry, and are often as accurate as state‐of‐the‐art electronic/atomistic models. Being an integral part of the ML architecture, representation of a molecular entity, uniquely encoded, plays a crucial role to what extent an ML model would be accurately predicting the desired property. This review aims to demonstrate a hierarchy of representations which has been introduced, to capture all degrees of freedom of a molecule or an atom the best, to map the quantum mechanical properties. We discuss their diverse applications how …
A Model of Graph Transactional Coverage Patterns with Applications to Drug Discovery
Srinivas Reddy Annappalli,Krishna Reddy Polepalli,Anirban Mondal,Deva Priyakumar U
International Conference on High Performance Computing, HiPC, 2021
@inproceedings{bib_A_Mo_2021, AUTHOR = {Annappalli, Srinivas Reddy and Polepalli, Krishna Reddy and Mondal, Anirban and U, Deva Priyakumar }, TITLE = {A Model of Graph Transactional Coverage Patterns with Applications to Drug Discovery}, BOOKTITLE = {International Conference on High Performance Computing}. YEAR = {2021}}
Facilitating the discovery of drugs by combining diverse compounds is becoming prevalent, especially for treating complex diseases like cancers and HIV. A drug is a chemi- cal compound structure and any sub-structure of a chemical compound is designated as a fragment. A chemical compound or a fragment can be modeled as a graph structure. Given a set of chemical compounds and their corresponding large set of fragments modeled as graph structures, we address the problem of identifying potential combinations of diverse chemical compounds, which cover a certain percentage of the set of fragments. In this regard, the key contributions of this work are three-fold: First, we introduce the notion of Graph Transactional Coverage Patterns (GTCPs) for any given graph transactional dataset. Second, we propose an efficient model and framework for extracting GTCPs from a given graph transactional dataset. Third, we conduct an extensive performance study using three real datasets to demonstrate that it is indeed feasible to efficiently extract GTCPs using our proposed GTCP-extraction framework. We also demonstrate the effectiveness of the GTCP-extraction framework through a case study in computer-aided drug design. Index Terms—Graph mining, Graph transactions, Coverage patterns, Drug discovery
A Model of Graph Transactional Coverage Patterns with Applications to Drug Discovery
Srinivas Reddy Annappalli,Krishna Reddy Polepalli,Anirban Mondal,Deva Priyakumar U
International Conference on High Performance Computing, HiPC, 2021
@inproceedings{bib_A_Mo_2021, AUTHOR = {Annappalli, Srinivas Reddy and Polepalli, Krishna Reddy and Mondal, Anirban and U, Deva Priyakumar }, TITLE = {A Model of Graph Transactional Coverage Patterns with Applications to Drug Discovery}, BOOKTITLE = {International Conference on High Performance Computing}. YEAR = {2021}}
Facilitating the discovery of drugs by combining diverse compounds is becoming prevalent, especially for treating complex diseases like cancers and HIV. A drug is a chemical compound structure and any sub-structure of a chemical compound is designated as a fragment. A chemical compound or a fragment can be modeled as a graph structure. Given a set of chemical compounds and their corresponding large set of fragments modeled as graph structures, we address the problem of identifying potential combinations of diverse chemical compounds, which cover a certain percentage of the set of fragments. In this regard, the key contributions of this work are three-fold: First, we introduce the notion of Graph Transactional Coverage Patterns (GTCPs) for any given graph transactional dataset. Second, we propose an efficient model and framework for extracting GTCPs from a given graph transactional dataset. Third, we conduct an extensive performance study using three real datasets to demonstrate that it is indeed feasible to efficiently extract GTCPs using our proposed GTCP-extraction framework. We also demonstrate the effectiveness of the GTCP-extraction framework through a case study in computer-aided drug design. Index Terms—Graph mining, Graph transactions, Coverage patterns, Drug discovery
IMLE-Net: An Interpretable Multi-level Multi-channel Model for ECG Classification
Likith Reddy,Vivek Talwar,Shanmukh Alle,Bapiraju Surampudi,Deva Priyakumar U
International Conference on Systems, Man, and Cybernetics, SMC, 2021
@inproceedings{bib_IMLE_2021, AUTHOR = {Reddy, Likith and Talwar, Vivek and Alle, Shanmukh and Surampudi, Bapiraju and U, Deva Priyakumar }, TITLE = {IMLE-Net: An Interpretable Multi-level Multi-channel Model for ECG Classification}, BOOKTITLE = {International Conference on Systems, Man, and Cybernetics}. YEAR = {2021}}
Early detection of cardiovascular diseases is crucial for effective treatment and an electrocardiogram (ECG) is pivotal for diagnosis. The accuracy of Deep Learning based methods for ECG signal classification has progressed in recent years to reach cardiologist-level performance. In clinical settings, a cardiologist makes a diagnosis based on the standard 12-channel ECG recording. Automatic analysis of ECG recordings from a multiple-channel perspective has not been given enough attention, so it is essential to analyze an ECG recording from a multiple-channel perspective. We propose a model that leverages the multiple-channel information available in the standard 12-channel ECG recordings and learns patterns at the beat, rhythm, and channel level. The experimental results show that our model achieved a macro-averaged ROC-AUC score of 0.9216, mean accuracy of 88.85% and a maximum F1 score of 0.8057 on the PTB-XL dataset. The attention visualization results from the interpretable model are compared against the cardiologist’s guidelines to validate the correctness and usability
Mining subgraph coverage patterns from graph transactions
Srinivas Reddy Annappalli,Krishna Reddy Polepalli,Anirban Mondal,Deva Priyakumar U
International Journal of Data Science and Analytics, IJDSA, 2021
@inproceedings{bib_Mini_2021, AUTHOR = {Annappalli, Srinivas Reddy and Polepalli, Krishna Reddy and Mondal, Anirban and U, Deva Priyakumar }, TITLE = {Mining subgraph coverage patterns from graph transactions}, BOOKTITLE = {International Journal of Data Science and Analytics}. YEAR = {2021}}
Pattern mining from graph transactional data (GTD) is an active area of research with applications in the domains of bioinformatics, chemical informatics and social networks. Existing works address the problem of mining frequent subgraphs from GTD. However, the knowledge concerning the coverage aspect of a set of subgraphs is also valuable for improving the performance of several applications. In this regard, we introduce the notion of subgraph coverage patterns (SCPs). Given a GTD, a subgraph coverage pattern is a set of subgraphs subject to relative frequency, coverage and overlap constraints provided by the user. We propose the Subgraph ID-based Flat Transactional (SIFT) framework for the efficient extraction of SCPs from a given GTD. Our performance evaluation using three real datasets demonstrates that our proposed SIFT framework is indeed capable of efficiently extracting SCPs from GTD. Furthermore, we demonstrate the effectiveness of SIFT through a case study in computer-aided drug design.
Artificial intelligence: machine learning for chemical sciences
Akshaya Karthikeyan,Deva Priyakumar U
Journal of Chemical Sciences, JCS, 2021
@inproceedings{bib_Arti_2021, AUTHOR = {Karthikeyan, Akshaya and U, Deva Priyakumar }, TITLE = {Artificial intelligence: machine learning for chemical sciences}, BOOKTITLE = {Journal of Chemical Sciences}. YEAR = {2021}}
Research in molecular sciences witnessed the rise and fall of Artificial Intelligence (AI)/ Machine Learning (ML) methods, especially artificial neural networks, few decades ago. However, we see a major resurgence in the use of modern ML methods in scientific research during the last few years. These methods have had phenomenal success in the areas of computer vision, speech recognition, natural language processing (NLP), etc. This has inspired chemists and biologists to apply these algorithms to problems in natural sciences. Availability of high performance Graphics Processing Unit (GPU) accelerators, large datasets, new algorithms, and libraries has enabled this surge. ML algorithms have successfully been applied to various domains in molecular sciences by providing much faster and sometimes more accurate solutions compared to traditional methods like Quantum Mechanical (QM) calculations, Density Functional Theory (DFT) or Molecular Mechanics (MM) based methods, etc. Some of the areas where the potential of ML methods are shown to be effective are in drug design, prediction of high–level quantum mechanical energies, molecular design, molecular dynamics materials, and retrosynthesis of organic compounds, etc. This article intends to conceptually introduce various modern ML methods and their relevance and applications in computational natural sciences.
MoleGuLAR: Molecule Generation using Reinforcement Learning with Alternating Rewards
Manan Goel,Shampa Raghunathan,Siddhartha Laghuvarapu,Deva Priyakumar U
Journal of Chemical Information and Modeling, JCIM, 2021
@inproceedings{bib_Mole_2021, AUTHOR = {Goel, Manan and Raghunathan, Shampa and Laghuvarapu, Siddhartha and U, Deva Priyakumar }, TITLE = {MoleGuLAR: Molecule Generation using Reinforcement Learning with Alternating Rewards}, BOOKTITLE = {Journal of Chemical Information and Modeling}. YEAR = {2021}}
The design of new inhibitors for novel targets is a very important problem especially in the current scenario with the world being plagued by COVID-19. Conventional approaches such as high-throughput virtual screening require extensive combing through existing datasets in the hope of finding possible matches. In this study, we propose a computational strategy for de novo generation of molecules with high binding affinities to the specified target and other desirable properties for drug-like molecules using reinforcement learning. A deep generative model built using a stack-augmented recurrent neural network initially trained to generate drug-like molecules is optimized using reinforcement learning to start generating molecules with desirable properties like LogP, Quantitative Estimate of Drug Likeliness, Topological Polar Surface Area, and Hydration Free Energy along with the binding affinity. For multi-objective optimization, we have devised a novel strategy in which the property being used to calculate the
SCONES: Self-Consistent Neural Network for Protein Stability Prediction Upon Mutation
Yashas Samaga B L,Shampa Raghunathan,Deva Priyakumar U
The Journal of Physical Chemistry B, JPCB, 2021
@inproceedings{bib_SCON_2021, AUTHOR = {L, Yashas Samaga B and Raghunathan, Shampa and U, Deva Priyakumar }, TITLE = {SCONES: Self-Consistent Neural Network for Protein Stability Prediction Upon Mutation}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2021}}
Engineering proteins to have desired properties by mutating amino acids at specific sites is commonplace. Such engineered proteins must be stable to function. Experimental methods used to determine stability at throughputs required to scan the protein sequence space thoroughly are laborious. To this end, many machine learning based methods have been developed to predict thermodynamic stability changes upon mutation. These methods have been evaluated for symmetric consistency by testing with hypothetical reverse mutations. In this work, we propose transitive data augmentation, evaluating transitive consistency with our new Stransitive data set, and a new machine learning based method, the first of its kind, that incorporates both symmetric and transitive properties into the architecture. Our method, called SCONES, is an interpretable neural network that predicts small relative protein stability changes for missense mutations that do not significantly alter the structure. It estimates a residue’s contributions toward protein stability (ΔG) in its local structural environment, and the difference between independently predicted contributions of the reference and mutant residues is reported as ΔΔG. We show that this self-consistent machine learning architecture is immune to many common biases in data sets, relies less on data than existing methods, is robust to overfitting, and can explain a substantial portion of the variance in experimental data.
Stereomutation in Tetracoordinate Centers via Stabilization of Planar Tetracoordinated Systems
Komal Yadav ,Upakarasamy Lourderaj,Deva Priyakumar U
Applied Theory On Molecular Systems, ATOMS, 2021
@inproceedings{bib_Ster_2021, AUTHOR = {, Komal Yadav and Lourderaj, Upakarasamy and U, Deva Priyakumar }, TITLE = {Stereomutation in Tetracoordinate Centers via Stabilization of Planar Tetracoordinated Systems}, BOOKTITLE = {Applied Theory On Molecular Systems}. YEAR = {2021}}
The quest for stabilizing planar forms of tetracoordinate carbon started five decades ago and intends to achieve interconversion between [R]- and [S]-stereoisomers without breaking covalent bonds. Several strategies are successful in making the planar tetracoordinate form a minimum on its potential energy surface. However, the first examples of systems where stereomutation is possible were reported only recently. In this study, the possibility of neutral and dications of simple hydrocarbons (cyclopentane, cyclopentene, spiropentane, and spiropentadiene) and their counterparts with the central carbon atom replaced by elements from groups 13, 14, and 15 are explored using ab initio MP2 calculations. The energy difference between the tetrahedral and planar forms decreases from row II to row III or IV substituents. Additionally, aromaticity involving the delocalization of the lone pair on the central atom appears to help in further stabilizing the planar form compared to the tetrahedral form, especially for the row II substituents. We identified 11 systems where the tetrahedral state is a minimum on the potential energy surface, and the planar form is a transition state corresponding to stereomutation. Interestingly, the planar structures of three systems were found to be minimum, and the corresponding tetrahedral states were transition states. The energy profiles corresponding to such transitions involving both planar and tetrahedral states without the breaking of covalent bonds were examined. The systems showcased in this study and research in this direction are expected to realize molecules that experimentally exhibit stereomutation.
Desolvation of Peptide Bond by O to S Substitution Impacts Protein Stability
Padmakar Wagh,Bhavesh Khatri,Shampa Raghunathan,Sohini Chakraborty,R Rahisuddin,Sangaralingam Kumaran,Raghu Tadala,Deva Priyakumar U,Jayanta Chatterjee
Angewandte Chemie International Edition, ACIE, 2021
@inproceedings{bib_Deso_2021, AUTHOR = {Wagh, Padmakar and Khatri, Bhavesh and Raghunathan, Shampa and Chakraborty, Sohini and Rahisuddin, R and Kumaran, Sangaralingam and Tadala, Raghu and U, Deva Priyakumar and Chatterjee, Jayanta }, TITLE = {Desolvation of Peptide Bond by O to S Substitution Impacts Protein Stability}, BOOKTITLE = {Angewandte Chemie International Edition}. YEAR = {2021}}
The step-by-step workflow adopted for the analysis of water molecules in the binding sites of small molecules of our interest bound to protein crystal structures. (Note: We examined the local water network around O and S atoms of carbonyl (>C=O) and thiocarbonyl (>C=S) in small molecule ligands that are non-covalently bound to protein crystal structures (≤2.5Å) deposited in the Protein Data Bank (PDB). To minimize the influence of heterogeneous chemical micro-environment on the hydration profile of O and S atoms, we chose to confine our analyses to ligands that possess both carbonyl and thiocarbonyl moieties (in thioamides, thioureas, thiocarbamates, and dithiocarbamates). A subset of protein-ligand pairs where the ligand has at least one water molecule in its First Solvation Shell (FSS is defined as a sphere containing any point that lies within 4.5 Å of a ligand atom) was selected for further analysis.
Linear Prediction Residual for Efficient Diagnosis of Parkinson’s Disease from Gait
Shanmukh Alle,Deva Priyakumar U
International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI, 2021
@inproceedings{bib_Line_2021, AUTHOR = {Alle, Shanmukh and U, Deva Priyakumar }, TITLE = {Linear Prediction Residual for Efficient Diagnosis of Parkinson’s Disease from Gait}, BOOKTITLE = {International Conference on Medical Image Computing and Computer Assisted Intervention}. YEAR = {2021}}
Parkinson’s Disease (PD) is a chronic and progressive neurological disorder that results in rigidity, tremors and postural instability. There is no definite medical test to diagnose PD and diagnosis is mostly a clinical exercise. Although guidelines exist, about 10-30% of the patients are wrongly diagnosed with PD. Hence, there is a need for an accurate, unbiased and fast method for diagnosis. In this study, we propose LPGNet, a fast and accurate method to diagnose PD from gait. LPGNet uses Linear Prediction Residuals (LPR) to extract discriminating patterns from gait recordings and then uses a 1D convolution neural network with depth-wise separable convolutions to perform diagnosis. LPGNet achieves an AUC of 0.91 with a 21 times speedup and about 99% lesser parameters in the model compared to the state of the art. We also undertake an analysis of various cross-validation strategies used in literature in PD diagnosis from gait and find that most methods are affected by some form of data leakage between various folds which leads to unnecessarily large models and inflated performance due to overfitting. The analysis clears the path for future works in correctly evaluating their methods.
Clinico-Genomic Analysis Reveals Mutations Associated with COVID-19 Disease Severity: Possible Modulation by RNA Structure
Priyanka Mehta,Shanmukh Alle,Anusha Chaturvedi,Aparna Swaminathan, Sheeba Saifi,Ranjeet Maurya,Partha Chattopadhyay,Deva Priyakumar U,Vinod Palakkad Krishnanunni
pathogens, pg, 2021
@inproceedings{bib_Clin_2021, AUTHOR = {Mehta, Priyanka and Alle, Shanmukh and Chaturvedi, Anusha and Swaminathan, Aparna and Saifi, Sheeba and Maurya, Ranjeet and Chattopadhyay, Partha and U, Deva Priyakumar and Krishnanunni, Vinod Palakkad }, TITLE = {Clinico-Genomic Analysis Reveals Mutations Associated with COVID-19 Disease Severity: Possible Modulation by RNA Structure}, BOOKTITLE = {pathogens}. YEAR = {2021}}
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) manifests a broad spectrum of clinical presentations, varying in severity from asymptomatic to mortality. As the viral infection spread, it evolved and developed into many variants of concern. Understanding the impact of mutations in the SARS-CoV-2 genome on the clinical phenotype and associated co-morbidities is important for treatment and preventionas the pandemic progresses. Based on the mild, moderate, and severe clinical phenotypes, we analyzed the possible association between both, the clinical sub-phenotypes and genomic mutations with respect to the severity and outcome of the patients. We found a significant association between the requirement of respiratory support and co-morbidities. We also identified six SARS-CoV-2 genome mutations that were significantly correlated with severity and mortality in our cohort. We examined structural alterations at the RNA and protein levels as a result of three of these mutations: A26194T, T28854T, and C25611A, present in the Orf3a and N protein. The RNA secondary structure change due to the above mutations can be one of the modulators of the disease outcome. Our findings highlight the importance of integrative analysis in which clinical and genetic components of the disease are co-analyzed. In combination with genomic surveillance, the clinical outcome-associated mutations could help identify individuals for priority medical support.
Host metabolic reprogramming in response to SARS-CoV-2 infection: A systems biology approach
Sai Teja Reddy Moolamalla,Rami B,Ruchi Chauhan,Deva Priyakumar U,Vinod Palakkad Krishnanunni
Microbial Pathogenesis, MP, 2021
@inproceedings{bib_Host_2021, AUTHOR = {Moolamalla, Sai Teja Reddy and B, Rami and Chauhan, Ruchi and U, Deva Priyakumar and Krishnanunni, Vinod Palakkad }, TITLE = {Host metabolic reprogramming in response to SARS-CoV-2 infection: A systems biology approach}, BOOKTITLE = {Microbial Pathogenesis}. YEAR = {2021}}
Understanding the pathogenesis of SARS-CoV-2 is essential for developing effective treatment strategies. Viruses hijack the host metabolism to redirect the resources for their replication and survival. The influence of SARSCoV-2 on host metabolism is yet to be fully understood. In this study, we analyzed the transcriptomic data obtained from different human respiratory cell lines and patient samples (nasopharyngeal swab, peripheral blood mononuclear cells, lung biopsy, bronchoalveolar lavage fluid) to understand metabolic alterations in response to SARS-CoV-2 infection. We explored the expression pattern of metabolic genes in the comprehensive genomescale network model of human metabolism, Recon3D, to extract key metabolic genes, pathways, and reporter metabolites under each SARS-CoV-2-infected condition. A SARS-CoV-2 core metabolic interactome was constructed for network-based drug repurposing. Our analysis revealed the host-dependent dysregulation of glycolysis, mitochondrial metabolism, amino acid metabolism, nucleotide metabolism, glutathione metabolism, polyamine synthesis, and lipid metabolism. We observed different pro- and antiviral metabolic changes and generated hypotheses on how the host metabolism can be targeted for reducing viral titers and immunomodulation. These findings warrant further exploration with more samples and in vitro studies to test predictions.
APObind: A Dataset of Ligand Unbound Protein Conformations for Machine Learning Applications in De Novo Drug Design
Rishal Aggarwal,Akash Gupta,Deva Priyakumar U
Technical Report, arXiv, 2021
@inproceedings{bib_APOb_2021, AUTHOR = {Aggarwal, Rishal and Gupta, Akash and U, Deva Priyakumar }, TITLE = {APObind: A Dataset of Ligand Unbound Protein Conformations for Machine Learning Applications in De Novo Drug Design}, BOOKTITLE = {Technical Report}. YEAR = {2021}}
Protein-ligand complex structures have been utilised to design benchmark machine learning methods that perform important tasks related to drug design such as receptor binding site detection, small molecule docking and binding affinity prediction. However, these methods are usually trained on only ligand bound (or holo) conformations of the protein and therefore are not guaranteed to perform well when the protein structure is in its native unbound conformation (or apo), which is usually the conformation available for a newly identified receptor. A primary reason for this is that the local structure of the binding site usually changes upon ligand binding. To facilitate solutions for this problem, we propose a dataset called APObind that aims to provide apo conformations of proteins present in the PDBbind dataset, a popular dataset used in drug design. Furthermore, we explore the performance of methods specific to three use cases on this dataset, through which, the importance of validating them on the APObind dataset is demonstrated.
MMBERT: Multimodal BERT Pretraining for Improved Medical VQA
Yash Khare,Viraj Bagal,MINESH MATHEW,Adithi Devi,Deva Priyakumar U,Jawahar C V
IEEE International Symposium on Biomedical Imaging, ISBI, 2021
@inproceedings{bib_MMBE_2021, AUTHOR = {Khare, Yash and Bagal, Viraj and MATHEW, MINESH and Devi, Adithi and U, Deva Priyakumar and V, Jawahar C }, TITLE = {MMBERT: Multimodal BERT Pretraining for Improved Medical VQA}, BOOKTITLE = {IEEE International Symposium on Biomedical Imaging}. YEAR = {2021}}
Images in the medical domain are fundamentally different from the general domain images. Consequently, it is infeasible to directly employ general domain Visual Question Answering (VQA) models for the medical domain. Additionally, medical image annotation is a costly and time-consuming process. To overcome these limitations, we propose a solution inspired by self-supervised pretraining of Transformer-style architectures for NLP, Vision, and Language tasks. Our method involves learning richer medical image and text semantic representations using Masked Vision-Language Modeling as the pretext task on a large medical image+caption dataset. The proposed solution achieves new state-of-the-art performance on two VQA datasets for radiology images – VQA-Med 2019 and VQA-RAD, outperforming even the ensemble models of previous best solutions. Moreover, our solution provides attention maps which help in model interpretability.
MEMES: Machine learning framework for Enhanced MolEcular Screening
Sarvesh Mehta,Siddhartha Laghuvarapu,P YASHASWI,Aaftaab Sethi,Mallika Alvala,Deva Priyakumar U
Chemical Science, Chem Sci, 2021
@inproceedings{bib_MEME_2021, AUTHOR = {Mehta, Sarvesh and Laghuvarapu, Siddhartha and YASHASWI, P and Sethi, Aaftaab and Alvala, Mallika and U, Deva Priyakumar }, TITLE = {MEMES: Machine learning framework for Enhanced MolEcular Screening}, BOOKTITLE = {Chemical Science}. YEAR = {2021}}
In drug discovery applications, high throughput virtual screening exercises are routinely performed to determine an initial set of candidate molecules referred to as “hits”. In such an experiment, each molecule from a large small-molecule drug library is evaluated in terms of physical properties such as the docking score against a target receptor. In real-life drug discovery experiments, drug libraries are extremely large but still there is only a minor representation of the essentially infinite chemical space, and evaluation of physical properties for each molecule in the library is not computationally feasible. In the current study, a novel Machine learning framework for Enhanced MolEcular Screening (MEMES) based on Bayesian optimization is proposed for efficient sampling of the chemical space. The proposed framework is demonstrated to identify 90% of the top-1000 molecules from a molecular library of size about 100 million, while calculating the docking score only for about 6% of the complete library. We believe that such a framework would tremendously help to reduce the computational effort in not only drug-discovery but also areas that require such high-throughput experiments.
DeepPocket: Ligand Binding Site Detection and Segmentation using 3D Convolutional Neural Networks
Rishal Aggarwal, Akash Gupta,Vineeth Ravindra Chelur,Jawahar C V,Deva Priyakumar U
Journal of Chemical Information and Modeling, JCIM, 2021
@inproceedings{bib_Deep_2021, AUTHOR = {Aggarwal, Rishal and Gupta, Akash and Chelur, Vineeth Ravindra and V, Jawahar C and U, Deva Priyakumar }, TITLE = {DeepPocket: Ligand Binding Site Detection and Segmentation using 3D Convolutional Neural Networks}, BOOKTITLE = {Journal of Chemical Information and Modeling}. YEAR = {2021}}
A structure-based drug design pipeline involves the development of potential drug molecules or ligands that form stable complexes with a given receptor at its binding site. A prerequisite to this is finding druggable and functionally relevant binding sites on the 3D structure of the protein. Although several methods for detecting binding sites have been developed beforehand, a majority of them surprisingly fail in the identification and ranking of binding sites accurately. The rapid adoption and success of deep learning algorithms in various sections of structural biology beckons the usage of such algorithms for accurate binding site detection. As a combination of geometry based software and deep learning, we report a novel framework, DeepPocket that utilises 3D convolutional neural networks for the rescoring of pockets identified by Fpocket and further segments these identified cavities on the protein surface. Apart from this, we also propose another dataset SC6K containing protein structures submitted in the Protein Data Bank (PDB) from 1st January, 2018 till 28th February, 2020 for ligand binding site (LBS) detection. DeepPocket’s results on various binding site datasets and SC6K highlights its better
BiRDS - Binding Residue Detection fromProtein Sequences using Deep ResNets
Vineeth Ravindra Chelur,Deva Priyakumar U
Journal of Chemical Information and Modeling, JCIM, 2021
@inproceedings{bib_BiRD_2021, AUTHOR = {Chelur, Vineeth Ravindra and U, Deva Priyakumar }, TITLE = {BiRDS - Binding Residue Detection fromProtein Sequences using Deep ResNets}, BOOKTITLE = {Journal of Chemical Information and Modeling}. YEAR = {2021}}
Protein-drug interactions play important roles in many biological processes and therapeutics. Prediction of the active binding site of a protein helps discover and optimise these interactions leading to the design of better ligand molecules. The tertiary structure of a protein determines the binding sites available to the drug molecule. A quick and accurate prediction of the binding site from sequence alone without utilising the three-dimensional structure is challenging. Deep Learning has been used in a variety of biochemical tasks and has been hugely successful. In this paper, a Residual Neural Network (leveraging skip connections) is implemented to predict a protein's most active binding site. An Annotated Database of Druggable Binding Sites from the Protein DataBank, sc-PDB, is used for training the network. Features extracted from the Multiple Sequence Alignments (MSAs) of the protein generated using DeepMSA, such as Position-Speci c Scoring Matrix (PSSM), Secondary Structure (SS3), and Relative Solvent Accessibility (RSA), are provided as input to the network. A weighted binary cross-entropy loss function is used to counter the substantial imbalance in the
Ion Selectivity and Permeation Mechanism in a Cyclodextrin-based Channel
MUSUNURU PRATYUSHA,Siladitya Padhi,Deva Priyakumar U
The Journal of Physical Chemistry B, JPCB, 2021
@inproceedings{bib_Ion__2021, AUTHOR = {PRATYUSHA, MUSUNURU and Padhi, Siladitya and U, Deva Priyakumar }, TITLE = {Ion Selectivity and Permeation Mechanism in a Cyclodextrin-based Channel}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2021}}
Synthetic ion channels are a promising technology in the medical and materials sciences because of their ability to conduct ions. Channels based on cyclodextrin, a cyclic oligomer of glucose, are of particular interest because of their non-toxicity and bio-compatibility. Using molecular dynamics-based free energy calculations, this study identifies cyclodextrin channel types that are best suited to serve as synthetic ion channels. Free energy profiles show that the connectivity in the channel determines whether the channel is cation-selective or anion-selective. Furthermore, the energy barrier for ion transport is governed by the number of glucose molecules constituting the cyclodextrin units of the channel. A detailed mechanism is proposed for ion transport through these channels. Findings from this study will aid in designing cyclodextrinbased channels that could be either cation-selective or anion-selective, by modifying the linkages of the channel or the number of glucose molecules in the cyclodextrin rings.
Learning Atomic Interactions through Solvation Free Energy Prediction Using Graph Neural Networks
Yashaswi Pathak,Sarvesh Mehta,Deva Priyakumar U
Journal of Chemical Information and Modeling, JCIM, 2021
@inproceedings{bib_Lear_2021, AUTHOR = {Pathak, Yashaswi and Mehta, Sarvesh and U, Deva Priyakumar }, TITLE = {Learning Atomic Interactions through Solvation Free Energy Prediction Using Graph Neural Networks}, BOOKTITLE = {Journal of Chemical Information and Modeling}. YEAR = {2021}}
Accuracy on Minnesota Solvation Database (MNSOL) There are classical and quantum-mechanical simulation studies that use the MNSOL as the reference database. We choose solvation model based on density(SMD) for comparison with our CIGIN Model. We also compared CIGIN Model with semi-empirical methods, pure COSMO, and COSMO-RS.
Multiscale Modeling of Wobble to Watson–Crick-Like Guanine–Uracil Tautomerization Pathways in RNA
Shreya Chandorkar,Shampa Raghunathan,JAGANADE TANASHREE SANTOSH,Deva Priyakumar U
International journal of molecular sciences, IJMS, 2021
@inproceedings{bib_Mult_2021, AUTHOR = {Chandorkar, Shreya and Raghunathan, Shampa and SANTOSH, JAGANADE TANASHREE and U, Deva Priyakumar }, TITLE = {Multiscale Modeling of Wobble to Watson–Crick-Like Guanine–Uracil Tautomerization Pathways in RNA}, BOOKTITLE = {International journal of molecular sciences}. YEAR = {2021}}
Energetically unfavorable Watson–Crick (WC)-like tautomeric forms of nucleobases are known to introduce spontaneous mutations, and contribute to replication, transcription, and translation errors. Recent NMR relaxation dispersion techniques were able to show that wobble (w) G•U mispair exists in equilibrium with the short-lived, low-population WC-like enolic tautomers. Presently, we have investigated the wG•U → WC-like enolic reaction pathway using various theoretical methods: quantum mechanics (QM), molecular dynamics (MD), and combined quantum mechanics/molecular mechanics (QM/MM). The previous studies on QM gas phase calculations were inconsistent with experimental data. We have also explored the environmental effects on the reaction energies by adding explicit water. While the QM-profile clearly becomes endoergic in the presence of water, the QM/MM-profile remains consistently endoergic in the presence and absence of water. Hence, by including microsolvation and QM/MM calculations, the experimental data can be explained. For the G•Uenol → Genol•U pathway, the latter appears to be energetically more favorable throughout all computational models. This study can be considered as a benchmark of various computational models of wG•U to WC-like tautomerization pathways with and without the environmental effects, and may contribute on further studies of other mispairs as well.
Machine learning based clinical decision support system for early COVID-19 mortality prediction
Akshaya Karthikeyan,Akshit Garg,Vinod Palakkad Krishnanunni,Deva Priyakumar U
Frontiers in public health, FPH, 2021
@inproceedings{bib_Mach_2021, AUTHOR = {Karthikeyan, Akshaya and Garg, Akshit and Krishnanunni, Vinod Palakkad and U, Deva Priyakumar }, TITLE = {Machine learning based clinical decision support system for early COVID-19 mortality prediction}, BOOKTITLE = {Frontiers in public health}. YEAR = {2021}}
The coronavirus disease 2019 (COVID-19), caused by the virus SARS-CoV-2, is an acute respiratory disease that has been classified as a pandemic by the World Health Organization (WHO). The sudden spike in the number of infections and high mortality rates have put immense pressure on the public healthcare systems. Hence, it is crucial to identify the key factors for mortality prediction to optimize patient treatment strategy. Different routine blood test results are widely available compared to other forms of data like X-rays, CT-scans, and ultrasounds for mortality prediction. This study proposes machine learning (ML) methods based on blood tests data to predict COVID-19 mortality risk. A powerful combination of five features: neutrophils, lymphocytes, lactate dehydrogenase (LDH), high-sensitivity C-reactive protein (hs-CRP), and age helps to predict mortality with 96% accuracy. Various ML models (neural networks, logistic regression, XGBoost, random forests, SVM, and decision trees) have been trained and performance compared to determine the model that achieves consistently high accuracy across the days that span the disease. The best performing method using XGBoost feature importance and neural network classification, predicts with an accuracy of 90% as early as 16 days before the outcome. Robust testing with three cases based on days to outcome confirms the strong predictive performance and practicality of the proposed model. A detailed analysis and identification of trends was performed using these key biomarkers to provide useful insights for intuitive application. This study provide solutions that would help accelerate the decision-making process in healthcare systems for focused medical treatments in an accurate, early, and reliable manner
Urea-water solvation of protein side chain models
JAGANADE TANASHREE SANTOSH,ADITYA CHATTOPADHYAY,Shampa Raghunathan,Deva Priyakumar U
Journal of Molecular Liquids, JML, 2020
@inproceedings{bib_Urea_2020, AUTHOR = {SANTOSH, JAGANADE TANASHREE and CHATTOPADHYAY, ADITYA and Raghunathan, Shampa and U, Deva Priyakumar }, TITLE = {Urea-water solvation of protein side chain models}, BOOKTITLE = {Journal of Molecular Liquids}. YEAR = {2020}}
Aqueous urea stabilizes the unfolded states of protein due to their ability to solvate both hydrophilic and hydrophobic residues favorably. The nature of interactions that stabilize different types of amino acid side chains in their solvent exposed state is still not understood. To gain insights into the molecular level details of urea interactions with proteins in their unfolded states, we have performed atomistic molecular dynamics simulations and free energy calculations using the thermodynamic integration method on model systems representing side chains of all amino acids in different solvent environments (water and varying concentrations of aqueous urea). A systematic analysis of structural, energetic and dynamic parameters has been done to understand the detailed atomistic mechanism. The main aim of the current study is to unravel the nature of urea-amino acid interactions by emphasizing on the chemical nature …
Selectivity and transport in aquaporins from molecular simulation studies
Siladitya Padhi,Deva Priyakumar U
Vitamins & Hormones, V&H, 2020
@inproceedings{bib_Sele_2020, AUTHOR = {Padhi, Siladitya and U, Deva Priyakumar }, TITLE = {Selectivity and transport in aquaporins from molecular simulation studies}, BOOKTITLE = {Vitamins & Hormones}. YEAR = {2020}}
The transport of water through aquaporins is a dynamic process that involves rapid movement of a chain of water molecules through the pore of the aquaporin. Structures of aquaporins solved using X-ray crystallography have provided some insights into how water is transported through these channels, and how certain structural features of the pore might help exclude other solutes from passing through the pore. However, such techniques provide only a static picture, and a dynamic picture of the transport and selectivity mechanism at work in aquaporins is possible with molecular dynamics (MD) simulations. In MD simulations, the forces between the different atoms in a system are computed, and the atoms are then allowed to move under the influence of these forces. This allows the sampling of different conformations of the molecule being studied, including conformations that are crucial in driving biological phenomena like water transport. Simulation studies have provided insights into a number of aspects of aquaporins, including the role of the asparagine-proline-alanine (NPA) motif and the aromatic/arginine (ar/R) constriction, water transport mechanism, mechanisms defining the selectivity of the channel, interaction with lipids, response to external electric field, and binding of putative drug molecules. This chapter provides a brief review of the current status of computational modeling of aquaporins using MD simulations. Initially, a brief account of force fields and MD simulations is presented followed by an account of how MD simulations have contributed to further our understanding of different aspects of aquaporins.
Enantioseparation and chiral induction in Ag 29 nanoclusters with intrinsic chirality
Hiroto Yoshida,Masahiro Ehara,Deva Priyakumar U,Tsuyoshi Kawai,Takuya Nakashima
Chemical Science, Chem Sci, 2020
@inproceedings{bib_Enan_2020, AUTHOR = {Yoshida, Hiroto and Ehara, Masahiro and U, Deva Priyakumar and Kawai, Tsuyoshi and Nakashima, Takuya }, TITLE = {Enantioseparation and chiral induction in Ag 29 nanoclusters with intrinsic chirality}, BOOKTITLE = {Chemical Science}. YEAR = {2020}}
The optical activity of a metal nanocluster (NC) is induced either by an asymmetric arrangement of constituents or by a dissymmetric field of a chiral ligand layer. Herein, we unveil the origin of chirality in Ag29 NCs, which is attributed to the intrinsically chiral atomic arrangement. The X-ray crystal structure of a Ag29(BDT)12(TPP)4 NC (BDT: 1,3-benzenedithiol; TPP: triphenylphosphine) manifested the presence of intrinsic chirality in the outer shell capping the icosahedral achiral Ag13 core. The enantiomers of the Ag29(BDT)12(TPP)4 NC are separated by high-performance liquid chromatography (HPLC) using a chiral column for the first time, showing mirror-image circular dichroism (CD) spectra. The CD spectra are reproduced by time-dependent density functional theory (TDDFT) calculations based on enantiomeric Ag29 models with achiral 1,3-propanedithiolate ligands. The mechanism of chiral induction in the synthesis of Ag29(DHLA)12 (DHLA: a-dihydrolipoic acid) NCs with a chiral ligand system is further discussed with the aid of DFT calculations. The use of the enantiomeric DHLA ligand preferentially leads to a one-handed atomic arrangement which is more stable than the opposite one, inducing the enantiomeric excess in the population of intrinsically chiral Ag29 NCs with CD activity
Transition between [R]-and [S]-stereoisomers without bond breaking
Shampa Raghunathan,Komal Yadav,JAGANADE TANASHREE SANTOSH,VANGARA PRATHYUSHA,BIKKINA SWETHA,Upakarasamy Lourderaj,Deva Priyakumar U
Physical Chemistry Chemical Physics, PCCP, 2020
@inproceedings{bib_Tran_2020, AUTHOR = {Raghunathan, Shampa and Yadav, Komal and SANTOSH, JAGANADE TANASHREE and PRATHYUSHA, VANGARA and SWETHA, BIKKINA and Lourderaj, Upakarasamy and U, Deva Priyakumar }, TITLE = {Transition between [R]-and [S]-stereoisomers without bond breaking}, BOOKTITLE = {Physical Chemistry Chemical Physics}. YEAR = {2020}}
The fifty–year proposal of nondissociative racemization reaction of a tetracoordinated tetrahedral center from one enantiomer to another via a planar transition state by Hoffmann and coworkers has been explored by many research groups during the past five decades. A number of stable molecules with planar tetracoordinate and higher-coordinate centers have been designed and experimentally realized; however, there has not been a single example of molecular system that can possibly undergo such racemization. Here we show examples of molecular species that undergo inversion of stereochemistry around tetrahedral centers (Si, Al− and P+) either via a planar transition state or an intermediate state using quantum mechanical, ab initio quasi-classical dynamics calculations, and Born-Oppenheimer molecular dynamics (BOMD) simulations. This work is expected to provide potential leads for future studies on this fundamental phenomenon in chemistry.
Urea-aromatic interactions in biology
Shampa Raghunathan,JAGANADE TANASHREE SANTOSH,Deva Priyakumar U
Biophysical Reviews, BioR, 2020
@inproceedings{bib_Urea_2020, AUTHOR = {Raghunathan, Shampa and SANTOSH, JAGANADE TANASHREE and U, Deva Priyakumar }, TITLE = {Urea-aromatic interactions in biology}, BOOKTITLE = {Biophysical Reviews}. YEAR = {2020}}
Noncovalent interactions are key determinants in both chemical and biological processes. Among such processes, the hydrophobic interactions play an eminent role in folding of proteins, nucleic acids, formation of membranes, protein-ligand recognition, etc.. Though this interaction is mediated through the aqueous solvent, the stability of the above biomolecules can be highly sensitive to any small external perturbations, such as temperature, pressure, pH, or even cosolvent additives, like, urea-a highly soluble small organic molecule utilized by various living organisms to regulate osmotic pressure. A plethora of detailed studies exist covering both experimental and theoretical regimes, to understand how urea modulates the stability of biological macromolecules. While experimentalists have been primarily focusing on the thermodynamic and kinetic aspects, theoretical modeling predominantly involves mechanistic information at the molecular level, calculating atomistic details applying the force field approach to the high level electronic details using the quantum mechanical methods. The review focuses mainly on examples with biological relevance, such as (1) urea-assisted protein unfolding, (2) urea-assisted RNA unfolding, (3) urea lesion interaction within damaged DNA, (4) urea conduction through membrane proteins, and (5) protein-ligand interactions those explicitly address the vitality of hydrophobic interactions involving exclusively the urea-aromatic moiety.
Band nn: A deep learning framework for energy prediction and geometry optimization of organic small molecules
Siddhartha Laghuvarapu,Yashaswi Pathak,Deva Priyakumar U
Journal of Computational Chemistry, JCC, 2020
@inproceedings{bib_Band_2020, AUTHOR = {Laghuvarapu, Siddhartha and Pathak, Yashaswi and U, Deva Priyakumar }, TITLE = {Band nn: A deep learning framework for energy prediction and geometry optimization of organic small molecules}, BOOKTITLE = {Journal of Computational Chemistry}. YEAR = {2020}}
Recent advances in artificial intelligence along with the development of large data sets of energies calculated using quantum mechanical (QM)/density functional theory (DFT) methods have enabled prediction of accurate molecular energies at reasonably low computational cost. However, machine learning models that have been reported so far require the atomic positions obtained from geometry optimizations using high‐level QM/DFT methods as input in order to predict the energies and do not allow for geometry optimization. In this study, a transferable and molecule size‐independent machine learning model bonds (B), angles (A), nonbonded (N) interactions, and dihedrals (D) neural network (BAND NN) based on a chemically intuitive representation inspired by molecular mechanics force fields is presented. The model predicts the atomization energies of equilibrium and nonequilibrium structures as sum of energy contributions from bonds (B), angles (A), nonbonds (N), and dihedrals (D) at remarkable accuracy. The robustness of the proposed model is further validated by calculations that span over the conformational, configurational, and reaction space. The transferability of this model on systems larger than the ones in the data set is demonstrated by performing calculations on selected large molecules. Importantly, employing the BAND NN model, it is possible to perform geometry optimizations starting from nonequilibrium structures along with predicting their energies. © 2019 Wiley Periodicals, Inc.
Chemically interpretable graph interaction network for prediction of pharmacokinetic properties of drug-like molecules
Yashaswi Pathak,Siddhartha Laghuvarapu,Sarvesh Mehta,Deva Priyakumar U
American Association for Artificial Intelligence, AAAI, 2020
@inproceedings{bib_Chem_2020, AUTHOR = {Pathak, Yashaswi and Laghuvarapu, Siddhartha and Mehta, Sarvesh and U, Deva Priyakumar }, TITLE = {Chemically interpretable graph interaction network for prediction of pharmacokinetic properties of drug-like molecules}, BOOKTITLE = {American Association for Artificial Intelligence}. YEAR = {2020}}
Solubility of drug molecules is related to pharmacokinetic properties such as absorption and distribution, which affects the amount of drug that is available in the body for its action. Computational or experimental evaluation of solvation free energies of drug-like molecules/solute that quantify solubilities is an arduous task and hence development of reliable computationally tractable models is sought after in drug discovery tasks in pharmaceutical industry. Here, we report a novel method based on graph neural network to predict solvation free energies. Previous studies considered only the solute for solvation free energy prediction and ignored the nature of the solvent, limiting their practical applicability. The proposed model is an end-to-end framework comprising three phases namely, message passing, interaction and prediction phases. In the first phase, message passing neural network was used to compute inter-atomic interaction within both solute and solvent molecules represented as molecular graphs. In the interaction phase, features from the preceding step is used to calculate a solute-solvent interaction map, since the solvation free energy depends on how (un)favorable the solute and solvent molecules interact with each other. The calculated interaction map that captures the solute-solvent interactions along with the features from the message passing phase is used to predict the solvation free energies in the final phase. The model predicts solvation free energies involving a large number of solvents with high accuracy. We also show that the interaction map captures the electronic and steric factors that govern the solubility of drug-like molecules and hence is chemically interpretable.
Deep Learning Enabled Inorganic Material Generator
Yashaswi Pathak,Karandeep Singh Juneja,Girish Varma,Masahiro Ehara,Deva Priyakumar U
Physical Chemistry Chemical Physics, PCCP, 2020
@inproceedings{bib_Deep_2020, AUTHOR = {Pathak, Yashaswi and Juneja, Karandeep Singh and Varma, Girish and Ehara, Masahiro and U, Deva Priyakumar }, TITLE = {Deep Learning Enabled Inorganic Material Generator}, BOOKTITLE = {Physical Chemistry Chemical Physics}. YEAR = {2020}}
Recent years have witnessed utilization of modern machine learning approaches for predicting properties of material using available datasets. However, to identify potential candidates for material discovery, one has to systematically scan through a large chemical space and subsequently calculate the properties of all such samples. On the other hand, generative methods are capable of efficiently sampling the chemical space and can generate molecules/materials with desired properties. In this study, we report a deep learning based inorganic material generator (DING) framework consisting of a generator module and a predictor module. The generator module is developed based upon conditional variational autoencoders (CVAE) and the predictor module consists of three deep neural networks trained for predicting enthalpy of formation, volume per atom and energy per atom chosen to demonstrate the proposed method. The predictor and generator modules have been developed using a one hot key representation of the material composition. A series of tests were done to examine the robustness of the predictor models, to demonstrate the continuity of the latent material space, and its ability to generate materials exhibiting target property values. The DING architecture proposed in this paper can be extended to other properties based on which the chemical space can be efficiently explored for interesting materials/molecules.
Machine Learning for Accurate Force Calculations in Molecular Dynamics Simulations
PUNYASLOK PATTNAIK,Shampa Raghunathan,K TARUN TEJA,Prabhakar Bhimalapuram,Jawahar C V,Deva Priyakumar U
Journal of Physical Chemistry A, PCA, 2020
@inproceedings{bib_Mach_2020, AUTHOR = {PATTNAIK, PUNYASLOK and Raghunathan, Shampa and TEJA, K TARUN and Bhimalapuram, Prabhakar and V, Jawahar C and U, Deva Priyakumar }, TITLE = {Machine Learning for Accurate Force Calculations in Molecular Dynamics Simulations}, BOOKTITLE = {Journal of Physical Chemistry A}. YEAR = {2020}}
The computationally expensive nature of ab initio molecular dynamics simulations severely limits its ability to simulate large system sizes and long time scales, both of which are necessary to imitate experimental conditions. In this work, we explore an approach to make use of the data obtained using the quantum mechanical density functional theory (DFT) on small systems and use deep learning to subsequently simulate large systems by taking liquid argon as a test case. A suitable vector representation was chosen to represent the surrounding environment of each Ar atom, and a -NetFF machine learning model where, the neural network was trained to predict the di↵erence in resultant forces obtained by DFT and classical force fields was introduced. Molecular dynamics simulations were then performed using forces from the neural network for various system sizes and time scales depending on the properties we calculated. A comparison of properties obtained from the classical force field and the neural network model was presented alongside available experimental data to validate the proposed method.
Recent Advancements in Computing Reliable Binding Free Energies in Drug Discovery Projects
N. Arul Murugan,Vasanthanathan Poongavanam,Deva Priyakumar U
Structural Bioinformatics: Applications in Preclinical Drug Discovery Process, SB-APDDP, 2019
@inproceedings{bib_Rece_2019, AUTHOR = {Murugan, N. Arul and Poongavanam, Vasanthanathan and U, Deva Priyakumar }, TITLE = {Recent Advancements in Computing Reliable Binding Free Energies in Drug Discovery Projects}, BOOKTITLE = {Structural Bioinformatics: Applications in Preclinical Drug Discovery Process}. YEAR = {2019}}
In recent times, our healthcare system is being challenged by many drug-resistant microorganisms and ageing-associated diseases for which we do not have any drugs or drugs with poor therapeutic profile. With pharmaceutical technological advancements, increasing computational power and growth of related biomedical fields, there have been dramatic increase in the number of drugs approved in general, but still way behind in drug discovery for certain class of diseases. Now, we have access to bigger genomics database, better biophysical methods, and knowledge about chemical space with which we should be able to easily explore and predict synthetically feasible compounds for the lead optimization process. In this chapter, we discuss the limitations and highlights of currently available computational methods used for protein–ligand binding affinities estimation and this includes force-field, ab initio electronic structure theory and machine learning approaches. Since the electronic structure-based approach cannot be applied to systems of larger length scale, the free energy methods based on this employ certain approximations, and these have been discussed in detail in this chapter. Recently, the methods based on electronic structure theory and machine learning approaches also are successfully being used to compute protein–ligand binding affinities and other pharmacokinetic and pharmacodynamic properties and so have greater potential to take forward computer-aided drug discovery to newer heights
Computational modeling of the catalytic mechanism of hydroxymethylbilane synthase
NAVNEET BUNG,Arjit Roy,Deva Priyakumar U,Gopalakrishnan Bulusu
Physical Chemistry Chemical Physics, PCCP, 2019
@inproceedings{bib_Comp_2019, AUTHOR = {BUNG, NAVNEET and Roy, Arjit and U, Deva Priyakumar and Bulusu, Gopalakrishnan }, TITLE = {Computational modeling of the catalytic mechanism of hydroxymethylbilane synthase}, BOOKTITLE = {Physical Chemistry Chemical Physics}. YEAR = {2019}}
Hydroxymethylbilane synthase (HMBS), the third enzyme in the heme biosynthesis pathway, catalyzes the formation of 1-hydroxymethylbilane (HMB) by a stepwise polymerization of four molecules of porphobilinogen (PBG) using the dipyrromethane (DPM) cofactor. The mechanism by which HMBS polymerizes four units of PBG has not been elucidated to date. In vitro and in silico studies on HMBS have suggested certain residues with catalytic importance, but their specific role in the catalysis is unclear. To understand the catalytic mechanism of HMBS, quantum mechanical (QM) calculations were performed on model systems obtained from the active site of the human HMBS enzyme. The addition of one molecule of PBG to the DPM cofactor is carried out in four steps: (1) protonation of the substrate, PBG; (2) deamination of PBG; (3) electrophilic addition of the deaminated substrate to the terminal pyrrole ring of the enzyme-bound DPM cofactor and (4) deprotonation of the carbon atom at the a-position of the second ring of DPM. Based on the energy profiles from the QM calculations on cluster models, R26 is proposed to be the best suitable proton donor to the PBG moiety, which aids in the deamination of the substrate. During the electrophilic addition step, the intermediate formed is stabilized by the carboxylate side chain of the D99 residue. In the final deprotonation step, an extra proton from the second ring of DPM is transferred to R26 via the carboxylate side chain of D99, thus completing one cycle of the catalytic mechanism. The residues in the cluster model seem to play an important role in obtaining accurate energy barriers. All the stationary points along the reaction pathway have been characterized using QM calculations. The rate limiting step for the complete mechanism is found to be the deamination of the PBG moiety. The results of this study provide a detailed understanding of the catalytic mechanism and would help design future studies aimed at modulating the activity of HMBS
Energetic, structural and dynamic properties of nucleobase-urea interactions that aid in urea assisted RNA unfolding
JAGANADE TANASHREE SANTOSH,ADITYA CHATTOPADHYAY,Deva Priyakumar U
Scientific Reports, SR, 2019
@inproceedings{bib_Ener_2019, AUTHOR = {SANTOSH, JAGANADE TANASHREE and CHATTOPADHYAY, ADITYA and U, Deva Priyakumar }, TITLE = {Energetic, structural and dynamic properties of nucleobase-urea interactions that aid in urea assisted RNA unfolding}, BOOKTITLE = {Scientific Reports}. YEAR = {2019}}
Understanding the structure-function relationships of RNA has become increasingly important given the realization of its functional role in various cellular processes. Chemical denaturation of RNA by urea has been shown to be beneficial in investigating RNA stability and folding. Elucidation of the mechanism of unfolding of RNA by urea is important for understanding the folding pathways. In addition to studying denaturation of RNA in aqueous urea, it is important to understand the nature and strength of interactions of the building blocks of RNA. In this study, a systematic examination of the structural features and energetic factors involving interactions between nucleobases and urea is presented. Results from molecular dynamics (MD) simulations on each of the five DNA/RNA bases in water and eight different concentrations of aqueous urea, and free energy calculations using the thermodynamic integration method are presented. The interaction energies between all the nucleobases with the solvent environment and the transfer free energies become more favorable with respect to increase in the concentration of urea. Preferential interactions of urea versus water molecules with all model systems determined using Kirkwood-Buff integrals and two-domain models indicate preference of urea by nucleobases in comparison to water. The modes of interaction between urea and the nucleobases were analyzed in detail. In addition to the previously identified hydrogen bonding and stacking interactions between urea and nucleobases that stabilize the unfolded states of RNA in aqueous solution, NH-π interactions are proposed to be important. Dynamic properties of each of these three modes of interactions have been presented. The study provides fundamental insights into the nature of interaction of urea molecules with nucleobases and how it disrupts nucleic acids.
Comparative study of the efficiency of Au, Ag, Pd and Pt based mono and bimetallic trimer clusters for the CO oxidation reaction
SAUMYA GURTU, Sandhya Rai,Deva Priyakumar U
Journal of Indian Chemical Society, JICS, 2019
@inproceedings{bib_Comp_2019, AUTHOR = {GURTU, SAUMYA and Rai, Sandhya and U, Deva Priyakumar }, TITLE = {Comparative study of the efficiency of Au, Ag, Pd and Pt based mono and bimetallic trimer clusters for the CO oxidation reaction}, BOOKTITLE = {Journal of Indian Chemical Society}. YEAR = {2019}}
Understanding synergistic effect by calculating electronic structure is essential to fine-tune the catalytic properties of bimetallic nanoalloy clusters which might be used for design of novel efficient catalysts. Density functional theory PBE0 calculations were performed to investigate the structure and energetics of various intermediates involved in the CO oxidation reaction catalyzed by Au3-xYx (x= 0–3 and Y denotes Ag or, Pt or, Pd) trimeric clusters through two possible pathways: Eley-Rideal (ER) and Langmuir-Hinshelwood (LH). The results of this investigation show that the catalytic behavior of the nanocluster highly depends on its composition and the reaction site taken into consideration. The most active reaction centres of goldsilver, gold-palladium, gold-platinum clusters are gold, palladium and platinum atoms respectively. The gold-silver clusters and AuPt2 prefer ER mechanism whereas, gold-palladium and Au2Pt selectively favour LH mechanism in comparison to the other. Bimetallic clusters, in general, are more efficient in comparison to their pristine mono-metallic counterparts, in activating the OO bond for the reaction and have relatively easy CO2 dissociation. Overall results indicate that the alloyed clusters could potentially have a better catalytic activity as compared to pure gold clusters for CO oxidation at low temperatures.
Model molecules to classify CH⋯ O hydrogen-bonds
Amol M. Vibhute,Deva Priyakumar U,Arthi Ravi, Kana M. Sureshan
Chemical communications, C C, 2018
@inproceedings{bib_Mode_2018, AUTHOR = {Vibhute, Amol M. and U, Deva Priyakumar and Ravi, Arthi and Sureshan, Kana M. }, TITLE = {Model molecules to classify CH⋯ O hydrogen-bonds}, BOOKTITLE = {Chemical communications}. YEAR = {2018}}
We developed a set of conformationally locked molecules each of which makes a single CH⋯O H-bond/short contact and has different electron density at the acceptor oxygen atom. The downfield shift of the 1H NMR signals due to the hydrogen involved in the CH⋯O H-bond varied from 0.93–1.6 ppm, and the magnitude of Δδ is in correlation with the hybridization state of the acceptor oxygen and with the CH⋯O H-bond strengths quantified using a computational method.
Quantum mechanical investigation of the nature of nucleobase-urea stacking interaction, a crucial driving force in RNA unfolding in aqueous urea
NITISH ALODIA,JAGANADE TANASHREE SANTOSH,Deva Priyakumar U
Journal of Chemical Sciences, JCS, 2018
@inproceedings{bib_Quan_2018, AUTHOR = {ALODIA, NITISH and SANTOSH, JAGANADE TANASHREE and U, Deva Priyakumar }, TITLE = {Quantum mechanical investigation of the nature of nucleobase-urea stacking interaction, a crucial driving force in RNA unfolding in aqueous urea}, BOOKTITLE = {Journal of Chemical Sciences}. YEAR = {2018}}
Urea-assisted denaturation of protein and RNA has been shown to be a valuable tool to study their stabilities and folding phenomena. It has been shown that stacking interactions between nucleobases and urea are one of the driving forces of denaturation. In this study, the ability of urea to form unconventional stacking interactions with RNA bases is investigated by performing high-level quantum calculations (RI-MP2/aug-ccpVDZ level) on a few thousands of model systems. Four systems were considered based on the RNA nucleobases (GUA, ADE, CYT, and URA) for the investigation. For each system, a set of models were designed to study the role of hetero-atoms/groups of the nucleobases on stacking interactions with urea moiety with respect to every possible pair. Several plane-parallel complexes were generated with urea on top of aromatic systems to exhaustively study all possible factors for urea-nucleobases stacking interactions. Energy decomposition analysis (EDA), atoms in molecules (AIM) and natural bond orbital (NBO) analysis were performed to gain better insights on non-covalent stacking interactions. Dispersion component was found to be heavily stabilizing, while the EHF was found to be repulsive for all the four systems indicating lack of hydrogen bonding (HB) type interactions and presence of dispersion type interactions. Amide and carbonyl groups of urea molecule were found to play a major role in favourable stacking interactions. We demonstrate that along with functional groups present on the nucleobases, the orientation of urea molecules plays a vital role in stabilizing the urea-nucleobase non-covalent interactions. The proposed study quantifies and provides a comprehensive theoretical description of urea nucleobase unconventional stacking interactions which helps to unravel urea driven RNA unfolding mechanism.
Temperature dependence of the stability of ion pair interactions, and its implications on the thermostability of proteins from thermophiles
BIKKINA SWETHA,AGASTYA P BHATI,Siladitya Padhi,Deva Priyakumar U
Journal of Chemical Sciences, JCS, 2017
@inproceedings{bib_Temp_2017, AUTHOR = {SWETHA, BIKKINA and BHATI, AGASTYA P and Padhi, Siladitya and U, Deva Priyakumar }, TITLE = {Temperature dependence of the stability of ion pair interactions, and its implications on the thermostability of proteins from thermophiles}, BOOKTITLE = {Journal of Chemical Sciences}. YEAR = {2017}}
An understanding of the determinants of the thermal stability of thermostable proteins is expected to enable design of enzymes that can be employed in industrial biocatalytic processes carried out at high temperatures. A major factor that has been proposed to stabilize thermostable proteins is the high occurrence of salt bridges. The current study employs free energy calculations to elucidate the thermodynamics of the formation of salt bridge interactions and the temperature dependence, using acetate and methylguanidium ions as model systems. Three different orientations of the methylguanidinium approaching the carboxylate group have been considered for obtaining the free energy profiles. The association of the two ions becomes more favorable with an increase in temperature. The desolvation penalty corresponding to the association of the ion pair is the lowest at high temperatures. The occurrence of …
Ability of density functional theory methods to accurately model the reaction energy pathways of the oxidation of CO on gold cluster: A benchmark study
SAUMYA GURTU,SANDHYA RAI, Masahiro Ehara,Deva Priyakumar U
Theoretical Chemistry Accounts, TCA, 2016
@inproceedings{bib_Abil_2016, AUTHOR = {GURTU, SAUMYA and RAI, SANDHYA and Ehara, Masahiro and U, Deva Priyakumar }, TITLE = {Ability of density functional theory methods to accurately model the reaction energy pathways of the oxidation of CO on gold cluster: A benchmark study}, BOOKTITLE = {Theoretical Chemistry Accounts}. YEAR = {2016}}
Gold clusters are currently regarded as new-generation catalysts owing to their exceptional efficiency in accelerating several classes of reactions. Density functional theory (DFT) is the method of choice for the investigation of energy pathways of reactions assisted by metal nanoparticles due to their computational efficiency. However, the reliability of such theoretical studies depends to a large extent on the choice of the DFT functional used. In the present work, the performance of a series of DFT-based functionals to accurately model the prototypical CO oxidation reaction catalyzed by a cluster has been examined by comparing the results with those obtained from high-level ab initio CCSD(T) method. This comparison study has been carried along the two possible pathways [Eley–Rideal (ER) and the Langmuir–Hinshelwood (LH)]. No significant differences among the DFT functionals were observed in …
Dynamic ligand-based pharmacophore modeling and virtual screening to identify mycobacterial cyclopropane synthase inhibitors
CHINMAYEE CHOUDHURY,Deva Priyakumar U, G NARAHARI SASTR
Journal of Chemical Sciences, JCS, 2016
@inproceedings{bib_Dyna_2016, AUTHOR = {CHOUDHURY, CHINMAYEE and U, Deva Priyakumar and SASTR, G NARAHARI }, TITLE = {Dynamic ligand-based pharmacophore modeling and virtual screening to identify mycobacterial cyclopropane synthase inhibitors}, BOOKTITLE = {Journal of Chemical Sciences}. YEAR = {2016}}
Multidrug resistance in Mycobacterium tuberculosis (M. Tb) and its coexistence with HIV are the biggest therapeutic challenges in anti-M. Tb drug discovery. The current study reports a Virtual Screening (VS) strategy to identify potential inhibitors of Mycobacterial cyclopropane synthase (CmaA1), an important M. Tb target considering the above challenges. Five ligand-based pharmacophore models were generated from 40 different conformations of the cofactors of CmaA1 taken from molecular dynamics (MD) simulations trajectories of CmaA1. The screening abilities of these models were validated by screening 23 inhibitors and 1398 non-inhibitors of CmaA1. A VS protocol was designed with four levels of screening i.e., ligand-based pharmacophore screening, structure-based pharmacophore screening, docking and absorption, distribution, metabolism, excretion and the toxicity (ADMET) filters. In an …
Structure, interaction, and dynamics of Au/Pd bimetallic nanoalloys dispersed in aqueous ethylpyrrolidone, a monomeric moiety of polyvinylpyrrolidone
ADITI GUPTA,Bundet Boekfa,Hidehiro Sakurai,Masahiro Ehara,Deva Priyakumar U
The Journal of Physical Chemistry C, JPCC, 2016
@inproceedings{bib_Stru_2016, AUTHOR = {GUPTA, ADITI and Boekfa, Bundet and Sakurai, Hidehiro and Ehara, Masahiro and U, Deva Priyakumar }, TITLE = {Structure, interaction, and dynamics of Au/Pd bimetallic nanoalloys dispersed in aqueous ethylpyrrolidone, a monomeric moiety of polyvinylpyrrolidone}, BOOKTITLE = {The Journal of Physical Chemistry C}. YEAR = {2016}}
Bimetallic nanoparticles (NPs) have been shown to exhibit certain advantages over pure NPs in catalysis due to a synergistic effect. It is common to disperse NPs in a polymer matrix such as polyvinylpyrrolidone (PVP) to prevent flocculation, which imparts considerable electronic effects on the NPs. In the present study, the interactions between aqueous solutions of N-ethylpyrrolidone (EP, system chosen to model the monomeric form of PVP) and Au/Pd bimetallic NPs, which are relevant in catalysis, have been investigated using molecular dynamics simulations and density functional theory (DFT) method. The adequacy of the force fields used was assessed based on their ability to reproduce the structures and adsorption energies obtained using DFT calculations. The interactions of NPs with the environment were studied at various concentrations of aqueous solutions of EP to examine the strength of NP–EP and NP …
Urea–aromatic stacking and concerted urea transport: conserved mechanisms in urea transporters revealed by molecular dynamics
Siladitya Padhi,Deva Priyakumar U
Journal of Chemical Theory and Computation, JCTC, 2016
@inproceedings{bib_Urea_2016, AUTHOR = {Padhi, Siladitya and U, Deva Priyakumar }, TITLE = {Urea–aromatic stacking and concerted urea transport: conserved mechanisms in urea transporters revealed by molecular dynamics}, BOOKTITLE = {Journal of Chemical Theory and Computation}. YEAR = {2016}}
Urea transporters are membrane proteins that selectively allow urea molecules to pass through. It is not clear how these transporters allow rapid conduction of urea, a polar molecule, in spite of the presence of a hydrophobic constriction lined by aromatic rings. The current study elucidates the mechanism that is responsible for this rapid conduction by performing free energy calculations on the transporter dvUT with a cumulative sampling time of about 1.3 μs. A parallel arrangement of aromatic rings in the pore enables stacking of urea with these rings, which, in turn, lowers the energy barrier for urea transport. Such interaction of the rings with urea is proposed to be a conserved mechanism across all urea-conducting proteins. The free energy landscape for the permeation of multiple urea molecules reveals an interplay between interurea interaction and the solvation state of the urea molecules. This is for the first time …
Binding to Gold Nanocluster Alters the Hydrogen Bonding Interactions and Electronic Properties of Canonical and Size Expanded DNA Base Pairs
SANDHYA RAI,Harjinder Singh,Deva Priyakumar U
RSC Advances, RSCA, 2015
@inproceedings{bib_Bind_2015, AUTHOR = {RAI, SANDHYA and Singh, Harjinder and U, Deva Priyakumar }, TITLE = {Binding to Gold Nanocluster Alters the Hydrogen Bonding Interactions and Electronic Properties of Canonical and Size Expanded DNA Base Pairs}, BOOKTITLE = {RSC Advances}. YEAR = {2015}}
DNA molecules tagged to metal nanoparticles, especially gold nanoparticles (AuNPs), have been shown to exhibit potential applications in designing and fabrication of novel electronic nano-devices, but the binding mechanism between gold nanoparticles and DNA bases and its implications are not completely understood. In this work, a comprehensive study to examine the effect of structural perturbations offered to DNA base pairs in terms of size expansion and adsorption on a gold cluster (Au3) has been done using density functional theory. Geometric and electronic features of these complexes provide evidences for distortion of certain base pairs depending on the binding site of the cluster. This is further substantiated via normal mode, natural bond orbital (NBO) and atoms in molecules (AIM) analyses. The natural population analysis (NPA) and NBO analysis indicate that complexation greatly affects the charge distribution on the base pairs due to charge transfer between base pair and gold cluster. This charge redistribution may offer a possibility of higher order interactions. Upon complexation, a marked decrease in the HOMO-LUMO gap is observed, which is more profound in cases where size expanded bases are involved due to the extended π-conjugation of the fused benzine rings. This study demonstrates the possibility of combining structural modifications to DNA base pairs and subsequent binding to gold nano particles to modulate and achieve molecular systems with desired optico-electronic properties.
Double zipper helical assembly of deoxyoligonucleotides: mutual templating and chiral imprinting to form hybrid DNA ensembles
Nagarjun Narayanaswamy,Suresh Gorle,Deva Priyakumar U,T. Govindaraju
Chemical communications, C C, 2015
@inproceedings{bib_Doub_2015, AUTHOR = {Narayanaswamy, Nagarjun and Gorle, Suresh and U, Deva Priyakumar and Govindaraju, T. }, TITLE = {Double zipper helical assembly of deoxyoligonucleotides: mutual templating and chiral imprinting to form hybrid DNA ensembles}, BOOKTITLE = {Chemical communications}. YEAR = {2015}}
Herein, the conventional and unconventional hydrogen bonding potential of adenine in APA for double zipper helical assembly of deoxyoligonucleotides is demonstrated under ambient conditions. The quantum mechanical calculations supported the formation of hybrid DNA ensembles.
Binding to gold nanoclusters alters the hydrogen bonding interactions and electronic properties of canonical and size-expanded DNA base pairs
SANDHYA RAI,Harjinder Singh,Deva Priyakumar U
RSC Advances, RSCA, 2015
@inproceedings{bib_Bind_2015, AUTHOR = {RAI, SANDHYA and Singh, Harjinder and U, Deva Priyakumar }, TITLE = {Binding to gold nanoclusters alters the hydrogen bonding interactions and electronic properties of canonical and size-expanded DNA base pairs}, BOOKTITLE = {RSC Advances}. YEAR = {2015}}
DNA molecules tagged to metal nanoparticles, especially gold nanoparticles (AuNPs), have been shown to have potential applications in the design and fabrication of novel electronic nano-devices, but the binding mechanism between gold nanoparticles and DNA bases and its implications are not completely understood. In this work, a comprehensive study to examine the effect of structural perturbations caused to DNA base pairs in terms of size expansion and adsorption on a gold cluster (Au3) has been carried out using density functional theory. The geometric and electronic features of these complexes provide evidence for the distortion of certain base pairs depending on the binding site of the cluster. This is further substantiated via normal mode, natural bond orbital (NBO) and atoms in molecules (AIM) analyses. The natural population analysis (NPA) and NBO analysis indicate that complexation greatly affects …
Atomistic details of the molecular recognition of DNA-RNA hybrid duplex by ribonuclease H enzyme
Suresh Gorle,Deva Priyakumar U
Journal of Chemical Sciences, JCS, 2015
@inproceedings{bib_Atom_2015, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {Atomistic details of the molecular recognition of DNA-RNA hybrid duplex by ribonuclease H enzyme}, BOOKTITLE = {Journal of Chemical Sciences}. YEAR = {2015}}
Bacillus halodurans (Bh) ribonuclease H (RNase H) belongs to the nucleotidyl-transferase (NT) superfamily and is a prototypical member of a large family of enzymes that use two-metal ion (Mg2+ or Mn2+) catalysis to cleave nucleic acids. Long timescale molecular dynamics simulations have been performed on the BhRNase H-DNA-RNA hybrid complex and the respective monomers to understand the recognition mechanism, conformational preorganization, active site dynamics and energetics involved in the complex formation. Several structural and energetic analyses were performed and significant structural changes are observed in enzyme and hybrid duplex during complex formation. Hybrid molecule binding to RNase H enzyme leads to conformational changes in the DNA strand. The ability of the DNA strand in the hybrid duplex to sample conformations corresponding to typical A- and …
Molecular dynamics study of the structure, flexibility, and hydrophilicity of PETIM dendrimers: a comparison with PAMAM dendrimers
Subbarao Kanchi,Suresh Gorle,Deva Priyakumar U,K. G. Ayappa,Prabal K Maiti
The Journal of Physical Chemistry B, JPCB, 2015
@inproceedings{bib_Mole_2015, AUTHOR = {Kanchi, Subbarao and Gorle, Suresh and U, Deva Priyakumar and Ayappa, K. G. and Maiti, Prabal K }, TITLE = {Molecular dynamics study of the structure, flexibility, and hydrophilicity of PETIM dendrimers: a comparison with PAMAM dendrimers}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2015}}
A new class of dendrimers, the poly(propyl ether imine) (PETIM) dendrimer, has been shown to be a novel hyperbranched polymer having potential applications as a drug delivery vehicle. Structure and dynamics of the amine terminated PETIM dendrimer and their changes with respect to the dendrimer generation are poorly understood. Since most drugs are hydrophobic in nature, the extent of hydrophobicity of the dendrimer core is related to its drug encapsulation and retention efficacy. In this study, we carry out fully atomistic molecular dynamics (MD) simulations to characterize the structure of PETIM (G2–G6) dendrimers in salt solution as a function of dendrimer generation at different protonation levels. Structural properties such as radius of gyration (Rg), radial density distribution, aspect ratio, and asphericity are calculated. In order to assess the hydrophilicity of the dendrimer, we compute the number of bound …
Modeling the structure of SARS 3a transmembrane protein using a minimum unfavorable contact approach
K RAMAKRISHNA SYAMKISHOR,Siladitya Padhi,Deva Priyakumar U
Journal of Chemical Sciences, JCS, 2015
@inproceedings{bib_Mode_2015, AUTHOR = {SYAMKISHOR, K RAMAKRISHNA and Padhi, Siladitya and U, Deva Priyakumar }, TITLE = {Modeling the structure of SARS 3a transmembrane protein using a minimum unfavorable contact approach}, BOOKTITLE = {Journal of Chemical Sciences}. YEAR = {2015}}
3a is an accessory protein from SARS coronavirus that is known to play a significant role in the proliferation of the virus by forming tetrameric ion channels. Although the monomeric units are known to consist of three transmembrane (TM) domains, there are no solved structures available for the complete monomer. The present study proposes a structural model for the transmembrane region of the monomer by employing our previously tested approach, which predicts potential orientations of TM α-helices by minimizing the unfavorable contact surfaces between the different TM domains. The best model structure comprising all three α-helices has been subjected to MD simulations to examine its quality. The TM bundle was found to form a compact and stable structure with significant intermolecular interactions. The structural features of the proposed model of 3a account for observations from previous …
Sumoylation of Sir2 differentially regulates transcriptional silencing in yeast
Abdul Hannan,Neethu Maria Abraham,SIDDHARTH GOYAL,Imlitoshi Jamir,Deva Priyakumar U,Krishnaveni Mishra
Nucleic Acids Research, NAR, 2015
@inproceedings{bib_Sumo_2015, AUTHOR = {Hannan, Abdul and Abraham, Neethu Maria and GOYAL, SIDDHARTH and Jamir, Imlitoshi and U, Deva Priyakumar and Mishra, Krishnaveni }, TITLE = {Sumoylation of Sir2 differentially regulates transcriptional silencing in yeast}, BOOKTITLE = {Nucleic Acids Research}. YEAR = {2015}}
Silent information regulator 2 (Sir2), the founding member of the conserved sirtuin family of NAD+-dependent histone deacetylase, regulates several physiological processes including genome stability, gene silencing, metabolism and life span in yeast. Within the nucleus, Sir2 is associated with telomere clusters in the nuclear periphery and rDNA in the nucleolus and regulates gene silencing at these genomic sites. How distribution of Sir2 between telomere and rDNA is regulated is not known. Here we show that Sir2 is sumoylated and this modification modulates the intra-nuclear distribution of Sir2. We identify Siz2 as the key SUMO ligase and show that multiple lysines in Sir2 are subject to this sumoylation activity. Mutating K215 alone counteracts the inhibitory effect of Siz2 on telomeric silencing. SUMO modification of Sir2 impairs interaction with Sir4 but not Net1 and, furthermore, SUMO modified Sir2 shows …
DNA-RNA hybrid duplexes with decreasing pyrimidine content in the DNA strand provide structural snapshots for the A-to B-form conformational transition of nucleic acids
Suresh Gorle,Deva Priyakumar U
Physical Chemistry Chemical Physics, PCCP, 2014
@inproceedings{bib_DNA-_2014, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {DNA-RNA hybrid duplexes with decreasing pyrimidine content in the DNA strand provide structural snapshots for the A-to B-form conformational transition of nucleic acids}, BOOKTITLE = {Physical Chemistry Chemical Physics}. YEAR = {2014}}
DNA–RNA hybrids are heterogeneous nucleic acid duplexes consisting of a DNA strand and a RNA strand, and are formed as key intermediates in many important biological processes. They serve as substrates for the RNase H enzymatic activity, which has been exploited for several biomedical technologies such as antiviral and antisense therapies. To understand the relation of structural properties with the base composition in DNA–RNA hybrids, molecular dynamics (MD) simulations were performed on selected model systems by systematically varying the deoxypyrimidine (dPy) content from 0 to 100% in the DNA strand. The results suggest that the hybrid duplex properties are highly dependent on their deoxypyrimidine content of the DNA strand. However, such variations are not seen in their corresponding pure DNA and RNA duplex counterparts. It is also noticed that the systematic variation in deoxypyrimidine …
Modulation of structural, energetic and electronic properties of DNA and size-expanded DNA bases upon binding to gold clusters
SANDHYA RAI,SUPRIYA RANJAN,Harjinder Singh,Deva Priyakumar U
RSC Advances, RSCA, 2014
@inproceedings{bib_Modu_2014, AUTHOR = {RAI, SANDHYA and RANJAN, SUPRIYA and Singh, Harjinder and U, Deva Priyakumar }, TITLE = {Modulation of structural, energetic and electronic properties of DNA and size-expanded DNA bases upon binding to gold clusters}, BOOKTITLE = {RSC Advances}. YEAR = {2014}}
Gold cluster–nucleobase complexes have potential applications in designing and fabrication of novel electronic nano-devices, and there has been a surge in research activities in this area recently. Binding of gold clusters (Au3 and Au4) with DNA bases and size-expanded DNA bases (x-bases) have been studied using density functional theory employing high quality basis set. A comprehensive attempt has been made to examine several gold–nucleobase complexes with respect to change in the orientation of Au clusters with respect to all the titratable sites of the bases. Geometric and electronic features of these complexes provided evidences for existence of non-conventional hydrogen bonds, which was further substantiated via vibrational frequency and natural bond orbital (NBO) analysis. The nucleobases, both canonical and size-expanded forms, form stable complexes with both the gold clusters considered …
Atomistic detailed mechanism and weak cation-conducting activity of HIV-1 Vpu revealed by free energy calculations
Siladitya Padhi,RAGHUNATH REDDY,Shahid Jameel,Deva Priyakumar U
Plos One, Plos One, 2014
@inproceedings{bib_Atom_2014, AUTHOR = {Padhi, Siladitya and REDDY, RAGHUNATH and Jameel, Shahid and U, Deva Priyakumar }, TITLE = {Atomistic detailed mechanism and weak cation-conducting activity of HIV-1 Vpu revealed by free energy calculations}, BOOKTITLE = {Plos One}. YEAR = {2014}}
The viral protein U (Vpu) encoded by HIV-1 has been shown to assist in the detachment of virion particles from infected cells. Vpu forms cation-specific ion channels in host cells, and has been proposed as a potential drug target. An understanding of the mechanism of ion transport through Vpu is desirable, but remains limited because of the unavailability of an experimental structure of the channel. Using a structure of the pentameric form of Vpu – modeled and validated based on available experimental data – umbrella sampling molecular dynamics simulations (cumulative simulation time of more than 0.4 µs) were employed to elucidate the energetics and the molecular mechanism of ion transport in Vpu. Free energy profiles corresponding to the permeation of Na+ and K+ were found to be similar to each other indicating lack of ion selection, consistent with previous experimental studies. The Ser23 residue is shown to enhance ion transport via two mechanisms: creating a weak binding site, and increasing the effective hydrophilic length of the channel, both of which have previously been hypothesized in experiments. A two-dimensional free energy landscape has been computed to model multiple ion permeation, based on which a mechanism for ion conduction is proposed. It is shown that only one ion can pass through the channel at a time. This, along with a stretch of hydrophobic residues in the transmembrane domain of Vpu, explains the slow kinetics of ion conduction. The results are consistent with previous conductance studies that showed Vpu to be a weakly conducting ion channel.
Crenarcheal chromatin proteins Cren7 and Sul7 compact DNA by introducing rigid bends
Rosalie P.C. Driessen,He Meng,Suresh Gorle,Rajesh Shahapure,Deva Priyakumar U,Malcolm F. White,Helmut Schiessel,ohnvan Noort,Remus Th. Dame
Nucleic Acids Research, NAR, 2013
@inproceedings{bib_Cren_2013, AUTHOR = {Driessen, Rosalie P.C. and Meng, He and Gorle, Suresh and Shahapure, Rajesh and U, Deva Priyakumar and White, Malcolm F. and Schiessel, Helmut and Noort, ohnvan and Dame, Remus Th. }, TITLE = {Crenarcheal chromatin proteins Cren7 and Sul7 compact DNA by introducing rigid bends}, BOOKTITLE = {Nucleic Acids Research}. YEAR = {2013}}
Archaeal chromatin proteins share molecular and functional similarities with both bacterial and eukaryotic chromatin proteins. These proteins play an important role in functionally organizing the genomic DNA into a compact nucleoid. Cren7 and Sul7 are two crenarchaeal nucleoidassociated proteins, which are structurally homologous, but not conserved at the sequence level. Cocrystal structures have shown that these two proteins induce a sharp bend on binding to DNA. In this study, we have investigated the architectural properties of these proteins using atomic force microscopy, molecular dynamics simulations and magnetic tweezers. We demonstrate that Cren7 and Sul7 both compact DNA molecules to a similar extent. Using a theoretical model, we quantify the number of individual proteins bound to the DNA as a function of protein concentration and show that forces up to 3.5 pN do not affect this binding. Moreover, we investigate the flexibility of the bending angle induced by Cren7 and Sul7 and show that the protein–DNA complexes differ in flexibility from analogous bacterial and eukaryotic DNA-bending proteins.
Role of conformational properties on the transannular Diels–Alder reactivity of macrocyclic trienes with varying linker lengths
VANGARA PRATHYUSHA,Deva Priyakumar U
RSC Advances, RSCA, 2013
@inproceedings{bib_Role_2013, AUTHOR = {PRATHYUSHA, VANGARA and U, Deva Priyakumar }, TITLE = {Role of conformational properties on the transannular Diels–Alder reactivity of macrocyclic trienes with varying linker lengths}, BOOKTITLE = {RSC Advances}. YEAR = {2013}}
The effect of the linker length (–CH2–) connecting the diene and the dienophile in macrocyclic trienes on their transannular Diels–Alder (TADA) reactivity has been investigated using density functional theory (DFT) calculations. The relationship between the conformational properties of these reactants and their reaction energy barriers was examined and a quantitative relationship has been obtained. The transition state energy barriers were found to increase with an increase in the linker length, which is in contrast to the expected trend. The conformational preferences of the triene reactants were studied using replica exchange molecular dynamics (REMD) simulations. These calculations reveal that longer linkers lead to a decreased probability of the occurrence of conformations with the diene and dienophile parts of the system vicinal to each other, and thus lower reactivity of systems with long linkers. Excellent correlations between the transition state energies and the cumulative probabilities corresponding to a short distance between the reactive sites, along with the ability of the diene to exist in the s-cisoid form, were observed.
Structures, dynamics, and stabilities of fully modified locked nucleic acid (β-D-LNA and α-L-LNA) duplexes in comparison to pure DNA and RNA duplexes
Suresh Gorle,Deva Priyakumar U
The Journal of Physical Chemistry B, JPCB, 2013
@inproceedings{bib_Stru_2013, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {Structures, dynamics, and stabilities of fully modified locked nucleic acid (β-D-LNA and α-L-LNA) duplexes in comparison to pure DNA and RNA duplexes}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2013}}
Locked nucleic acid (LNA) is a chemical modification which introduces a -O-CH2-linkage in the furanose sugar of nucleic acids and blocks its conformation in a particular state. Two types of modifications, namely, 2'-O,4'-C-methylene-β-D-ribofuranose (β-D-LNA) and 2'-O,4'-C-methylene-α-L-ribofuranose (α-L-LNA), have been shown to yield RNA and DNA duplex-like structures, respectively. LNA modifications lead to increased melting temperatures of DNA and RNA duplexes, and have been suggested as potential therapeutic agents in antisense therapy. In this study, molecular dynamics (MD) simulations were performed on fully modified LNA duplexes and pure DNA and RNA duplexes sharing a similar sequence to investigate their structure, stabilities, and solvation properties. Both LNA duplexes undergo unwinding of the helical structure compared to the pure DNA and RNA duplexes. Though the α-LNA substituent has been proposed to mimic deoxyribose sugar in its conformational properties, the fully modified duplex was found to exhibit unique structural and dynamic properties with respect to the other three nucleic acid structures. Free energy calculations accurately capture the enhanced stabilization of the LNA duplex structures compared to DNA and RNA molecules as observed in experiments. π-stacking interaction between bases from complementary strands is shown to be one of the contributors to enhanced stabilization upon LNA substitution. A combination of two factors, namely, nature of the -O-CH2- linkage in the LNAs vs their absence in the pure duplexes and similar conformations of the sugar rings in DNA and α-LNA vs the other two, is suggested to contribute to the stark differences among the four duplexes studied here in terms of their structural, dynamic, and energetic properties
Computational investigation on the effect of thermal perturbation on the mechanical unfolding of Titin I27
NAVNEET BUNG,Deva Priyakumar U
Journal of Molecular Modeling, JMM, 2013
@inproceedings{bib_Comp_2013, AUTHOR = {BUNG, NAVNEET and U, Deva Priyakumar }, TITLE = {Computational investigation on the effect of thermal perturbation on the mechanical unfolding of Titin I27}, BOOKTITLE = {Journal of Molecular Modeling}. YEAR = {2013}}
The emergence of single-molecule force measurement experiments has facilitated a better understanding of protein folding pathways and the thermodynamics involved. Computational methods such as steered molecular dynamics (SMD) simulations are helpful in providing atomistic level information on the unfolding pathways. Recent experimental studies have showed that combinations of single-molecule experiments with traditional methods such as chemical and/or thermal denaturation yield additional insights into the folding phenomenon. In this study, we report results from extensive computations (a total of about 60 SMD simulations with a total length of about 0.4 μs) that address the effect of thermal perturbation on the mechanical stability of the I27 domain of the protein titin. A wide range of temperatures (280–340 K) were considered for the pulling, which was done at both constant velocity and constant force using SMD simulations. Good agreement with experimental data, such as for the trends in changes in average force and the maximum force with respect to the temperature, was obtained. This study identifies two competing pathways for the mechanical unfolding of I27, and illustrates the significance of combining various techniques to examine protein foldin
Role of hydrophobic core on the thermal stability of proteins – molecular dynamics simulations on a single point mutant of Sso7d abstract
Deva Priyakumar U
Journal of Biomolecular Structure and Dynamics, JBSD, 2013
@inproceedings{bib_Role_2013, AUTHOR = {U, Deva Priyakumar }, TITLE = {Role of hydrophobic core on the thermal stability of proteins – molecular dynamics simulations on a single point mutant of Sso7d abstract}, BOOKTITLE = {Journal of Biomolecular Structure and Dynamics}. YEAR = {2013}}
The role of salt bridges in chromatin protein Sso7d, from S. solfataricus has previously been shown to be crucial for its unusual high thermal stability. Experimental studies have shown that single site mutation of Sso7d (F31A) leads to a substantial decrease in the thermal stability of this protein due to distortion of the hydrophobic core. In the present study, we have performed a total of 0.2s long molecular dynamics (MD) simulations on F31A at room temperature, and at 360 K, close to the melting temperature of the wild type (WT) protein to investigate the role of hydrophobic core on protein stability. Sso7d-WT was shown to be stable at both 300 and 360 K; however, F31A undergoes denaturation at 360 K, consistent with experimental results. The structural and energetic properties obtained using the analysis of MD trajectories indicate that the single mutation results in high flexibility of the protein, and loosening of intramolecular interactions. Correlation between the dynamics of the salt bridges with the structural transitions and the unfolding pathway indicate the importance of both salt bridges and hydrophobic in effecting thermal stability of proteins in general.
Synthesis and Reactivity Studies of Dicationic Dihydrogen Complexes Bearing Sulfur-Donor Ligands: A Combined Experimental and Computational Study
Thirumanavelan Gandhi, S.Rajkumar,VANGARA PRATHYUSHA,Deva Priyakumar U
European Journal of Inorganic Chemistry, EJBM, 2013
@inproceedings{bib_Synt_2013, AUTHOR = {Gandhi, Thirumanavelan and S.Rajkumar, and PRATHYUSHA, VANGARA and U, Deva Priyakumar }, TITLE = {Synthesis and Reactivity Studies of Dicationic Dihydrogen Complexes Bearing Sulfur-Donor Ligands: A Combined Experimental and Computational Study}, BOOKTITLE = {European Journal of Inorganic Chemistry}. YEAR = {2013}}
A series of dihydrogen complexes trans-[Ru(η2-H2){SC(SR)H}(dppe)2][X][BF4] (R = CH3, X = OTf; R = C6H5CH2, X = BPh4; R = H2C=CHCH2, X = BPh4; dppe = Ph2PCH2CH2PPh2) bearing sulfur-donor ligands has been synthesized by protonation of the (alkyl dithioformate)hydrido complexes trans-[Ru(H){SC(SR)H}(dppe)2][X] by using HBF4•Et2O. Competitive substitution reactions between H2 and SC(SR)H in trans-[Ru(η2-H2){SC(SR)H}(dppe)2][X][BF4] have been studied by treatment with CH3CN, CO, and P(OCH3)3. These resulted in the expulsion of SC(SR)H from the metal center, thus indicating that the alkyl dithioformate ligand is more labile than H2. Bonding of alkyl dithioformate ligands (sulfur-donor ligands) trans to H2 have been studied by comparing the H–H distances and chemical-shift values (1H NMR spectroscopy) of the various dihydrogen complexes bearing different trans ligands. This study qualitatively suggests that the alkyl dithioformate ligands in these trans-dihydrogen complexes show a poor π effect, and it is further supported by density functional theory calculations. The first example of a dihydrogen complex bearing dithioformic acid, trans-[Ru(η2-H2){SC(SH)H}(dppe)2][BF4]2, was obtained by protonation of trans-[Ru(H){SC(S)H}(dppe)2] by using HBF4•Et2O.
An Anti- and Pro- van't Hoff-Le Bel Moiety: Computational Study of a Phosphonium Ion and Related Compounds
VANGARA PRATHYUSHA,Deva Priyakumar U
Indo-German Meeting on Modeling Chemical and Biological (Re)Activity, MCBR3, 2013
@inproceedings{bib_An_A_2013, AUTHOR = {PRATHYUSHA, VANGARA and U, Deva Priyakumar }, TITLE = {An Anti- and Pro- van't Hoff-Le Bel Moiety: Computational Study of a Phosphonium Ion and Related Compounds}, BOOKTITLE = {Indo-German Meeting on Modeling Chemical and Biological (Re)Activity}. YEAR = {2013}}
In the late 19th century, van’t Hoff and Le Bel proposed that carbon prefers a tetrahedral arrangement in its tetracoordinate form. Similarly, the isoelectronic species such as B-, N+, Al-, Si and P+ assume tetrahedral arrangement. However, it has been shown by several research groups that planar arrangement for tetracoordinated carbon centers may be possible for certain molecules that are stabilized by steric constraints and electronic effects. Sastry and coworkers have identified a series of neutral hydrocarbons that are stable in their planar form. We have done high level quantum chemical calculations on skeletally substituted derivative of one of the hydrocarbons (C5H4). Interestingly, the phosphonium ion (C4PH4 +) is found to be a minimum on its potential energy surface consistently by both density functional theory and MP2 methods. A transition state corresponding to the interconversion between the two forms was identified, which indicates rapid transitions. Additionally, the tetrahedral form was found to be the minima for C and B-. The tetrahedral form is the minima for Al- and Si, whereas the planar form was characterized as a transition state that connects two identical tetrahedral forms. Ab initio molecular dynamics calculations indicate that C4H4Si does undergo rapid interconversion between two identical tetrahedral structures through the planar transition state in the femtosecond timescale. However, C4H4P+ and C4H4Al- undergoes irreversible ring opening during the simulations. These molecules/ions were further analyzed using NICS calculations and ring opening reactivities.
Structure, Dynamics and Stability of Pure and Chemically Modified Nucleic acids Studied via Molecular Dynamics Simulations
Suresh Gorle,Deva Priyakumar U
Indo-German Meeting on Modeling Chemical and Biological (Re)Activity, MCBR3, 2013
@inproceedings{bib_Stru_2013, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {Structure, Dynamics and Stability of Pure and Chemically Modified Nucleic acids Studied via Molecular Dynamics Simulations}, BOOKTITLE = {Indo-German Meeting on Modeling Chemical and Biological (Re)Activity}. YEAR = {2013}}
Due to poor stability and nature of low nuclease resistance of pure nucleic acids, a suitable chemical modification is required which improves several factors such as increased cellular uptake, altered duplex conformation and stability. In general, the modifications at ribose sugar (backbone) of the oligonucleotide attribute special features to the duplexes than other modifications. Some of the common modifications include 2'-O-methyl, 2'-O,4'-C-methylene-β-D-ribofuranosyl (LNA) and phosphoroamidate. The modified oligonucleotides have also been shown to be potential antisense agents. Molecular dynamics (MD) simulations are valuable tools to study nucleic acid dynamics, conformational preferences, stabilities etc. Here, we have employed different simulation techniques such as traditional MD, replica exchange MD to understand the properties of modified duplexes and also the origin for the effect of chemical modification. The simulations revealed several important aspects related to duplex stability and relation with their base content. The factors responsible for the stability/unstability of modified duplexes with different sequences are also examined. Our results also suggest that the differential changes observed in nucleic acids depends on the position, nature of the modified site. This study also provides a molecular level picture of structural changes arising due to different types of modifications which is important in designing antisense oligonucleotides.
Elucidation of Oligomeric Structure and Ion Permeation Mechanism through the Viral Ion Channels HIV-1 Vpu and SARS 3a
Siladitya Padhi,K RAMAKRISHNA SYAMKISHOR,GOTTIPATI RAGINI,Deva Priyakumar U
Indo-German Meeting on Modeling Chemical and Biological (Re)Activity, MCBR3, 2013
@inproceedings{bib_Eluc_2013, AUTHOR = {Padhi, Siladitya and SYAMKISHOR, K RAMAKRISHNA and RAGINI, GOTTIPATI and U, Deva Priyakumar }, TITLE = {Elucidation of Oligomeric Structure and Ion Permeation Mechanism through the Viral Ion Channels HIV-1 Vpu and SARS 3a}, BOOKTITLE = {Indo-German Meeting on Modeling Chemical and Biological (Re)Activity}. YEAR = {2013}}
The structure and ion channel activity of the viral ion channels Vpu from human immunodeficiency virus type I (HIV-1) and 3a from SARS Coronavirus (SARS-CoV) have not been characterised in atomistic detail, in spite of the physiological importance of these ion channels. We have modelled several possible oligomeric states for Vpu, and carried out extensive molecular dynamics (MD) simulations in a lipid bilayer environment. Our results show that the pentamer is the most stable oligomeric form, with van der Waals interactions being the dominant force in holding together the monomeric units. The mechanism of ion conduction through the pentameric form has been investigated by umbrella sampling free energy calculations. We suggest that only one ion can pass through the channel at a time, rather than a continuous stream of ions; this is in agreement with experimental conductance studies [1]. We have also modelled possible structures for a monomeric unit of the ion channel SARS 3a, which has three helical transmembrane (TM) domains connected by loops [2]. Based on residue orientation, clusters of hydrophilic residues have been identified in each TM domain, and, by considering all possible orientations of these hydrophilic clusters, a number of monomers have been modelled and equilibrated in a membrane environment. We shall be presenting model structures of both monomeric and oligomeric states of 3a
Inter- versus intra-molecular cyclization of tripeptides containing tetrahydrofuran amino acids: a density functional theory study on kinetic control
Suresh Kumar N V,Deva Priyakumar U,Harjinder Singh,Saumya Roy,Tushar Kanti Chakraborty
Journal of Molecular Modeling, JMM, 2012
@inproceedings{bib_Inte_2012, AUTHOR = {V, Suresh Kumar N and U, Deva Priyakumar and Singh, Harjinder and Roy, Saumya and Chakraborty, Tushar Kanti }, TITLE = {Inter- versus intra-molecular cyclization of tripeptides containing tetrahydrofuran amino acids: a density functional theory study on kinetic control}, BOOKTITLE = {Journal of Molecular Modeling}. YEAR = {2012}}
: Density functional B3LYP method was used to investigate the preference of intra and intermolecular cyclizations of linear tripeptides containing tetrahydrofuran amino acids. Two distinct model pathways were conceived for the cyclization reaction and all possible transition states and intermediates were located. Analysis of the energetics indicate intermolecular cyclization being favored by both thermodynamic and kinetic control. Geometric and NBO analyses were performed to explain the trends obtained along both the reaction pathways. Conceptual density functional theory based reactive indices also show that reaction pathways leading to intermolecular cyclization of the tripeptides are relatively more facile compared to intramolecular cyclization.
Combining Thermal, Chemical and Mechanical Perturbations to Study Protein Folding: Molecular Dynamics Studies on Trp-cage Miniprotein, and Titin I27
KASAVAJHALA KOUSHIK,NAVNEET BUNG,Deva Priyakumar U
ACS Meeting, ACS-M, 2012
@inproceedings{bib_Comb_2012, AUTHOR = {KOUSHIK, KASAVAJHALA and BUNG, NAVNEET and U, Deva Priyakumar }, TITLE = {Combining Thermal, Chemical and Mechanical Perturbations to Study Protein Folding: Molecular Dynamics Studies on Trp-cage Miniprotein, and Titin I27}, BOOKTITLE = {ACS Meeting}. YEAR = {2012}}
t has been proposed that combining chemical and thermal denaturation along with mechanical unfolding enables better understanding of the thermodynamics of protein folding, and associated pathways. We have used molecular dynamics (MD) simulations, and umbrella sampling technique to study the effect of thermal and chemical perturbations on the unfolding of Trp-cage miniprotein (TC5b). Free energy profiles for the unfolding of the minicage protein were obtained at two different temperatures, and in presence and absence of urea. The distance between the penultimate terminal residues was used as the reaction coordinate for calculating the free energies. The unfolding involves opening of the Trp-cage, and separation of the secondary structure elements followed by the unfolding of the α-helix. Several energetic and geometric analyses including inter- and intramolecular interactions, and radial distribution functions and their relevance to the folding pathways will be discussed. Similar study on the titin I27 protein will also be presented.
Atomistic Details of the Molecular Recognition and Switching Mechanisms of SAM-III Riboswitch
HARINI SRINIVASAN,SHIVANI NANDA,Deva Priyakumar U
ACS Meeting, ACS-M, 2012
@inproceedings{bib_Atom_2012, AUTHOR = {SRINIVASAN, HARINI and NANDA, SHIVANI and U, Deva Priyakumar }, TITLE = {Atomistic Details of the Molecular Recognition and Switching Mechanisms of SAM-III Riboswitch}, BOOKTITLE = {ACS Meeting}. YEAR = {2012}}
SAM-III riboswitch present in the 5’-UTR of mRNA that translates to SAM synthetase in certain bacteria is involved in regulating the biosynthesis of S-adenosylmethionine (SAM) by acting as an on-off switch for the translation process. We have used molecular dynamics (MD) and steered MD simulations to understand the ligand recognition and the switching mechanisms. Computations on model systems indicated that the riboswitch does not recognize SAM over a structurally related analog, S-adenosylhomocysteine (SAH). Nonetheless, the RNA uses nonspecific interactions to stabilize SAM in its binding pocket compared to SAH. Calculations were done in the absence of the ligand to examine the conformational changes responsible for the transition between on and off states. While the on- state is capable of sampling a large conformational space, formation of duplexlike structure comprising the adenine part of the ligand and sequentially distant nucleotides seems to be crucial for the on- to off- state transition.
Determination of the Native Oligomeric State of Vpu, Transmembrane Protein from HIV-1, and its Ion Channel Activity
Siladitya Padhi,Harjinder Singh,Shahid Jameel,Deva Priyakumar U
Lipid-Protein Interactions in Membranes: Implications for Health and Disease, LPIM:IHD, 2012
@inproceedings{bib_Dete_2012, AUTHOR = {Padhi, Siladitya and Singh, Harjinder and Jameel, Shahid and U, Deva Priyakumar }, TITLE = {Determination of the Native Oligomeric State of Vpu, Transmembrane Protein from HIV-1, and its Ion Channel Activity}, BOOKTITLE = {Lipid-Protein Interactions in Membranes: Implications for Health and Disease}. YEAR = {2012}}
Vpu is an 81-residue protein encoded by human immunodeficiency virus type I (HIV-1) that facilitates viral release from host cells. The protein has a cytoplasmic domain and a helical transmembrane (TM) domain, of which the latter oligomerizes to form cation-specific ion channels. The number of TM domains that constitute the channel is still unclear, with experimental studies indicating the existence of a variety of oligomeric states. In this study, we have examined the possibility of Vpu to exist in tetra-, penta-, and hexameric states using comprehensive molecular dynamics (MD) simulations. By modeling the TM domain as an ideal α-helix, we carried out replica-exchange MD simulations in an implicit membrane environment for obtaining suitable starting structures, which were then subjected to extensive MD simulations in a fully hydrated lipid bilayer environment. The results show that the pentameric form is the most stable oligomeric state (the tetramer and hexamer models lose their initial channel-like structure), with helices in the pentamer being held together by strong van der Waals interactions. Hydrogen bonds between lipid headgroups and basic/hydrophilic residues on the protein are stronger in the pentamer than in the tetramer or the hexamer, indicating that these interactions might play a role in adhering the pentamer to the membrane. Free energy calculations using umbrella sampling technique have been performed to examine the potassium ion permeation through the pentameric pore. The results show a high free energy barrier corresponding to ion transport indicating weak ion channel activity in agreement with previous experimental biophysical studies.
Impact of chemical modification on the structures, dynamics and stability of DNA-RNA hybrids: A molecular dynamics simulation study
Suresh Gorle,Deva Priyakumar U
Theoritical Chemistry Symposium, TCS, 2012
@inproceedings{bib_Impa_2012, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {Impact of chemical modification on the structures, dynamics and stability of DNA-RNA hybrids: A molecular dynamics simulation study}, BOOKTITLE = {Theoritical Chemistry Symposium}. YEAR = {2012}}
Chemical modifications on the backbone of the oligonucleotides have been shown to improve several factors such as increased cellular uptake, altered duplex conformation and stability. Modified oligonucleotides have also been shown to be potential antisense agents. Some of the common modifications include 2'-O-methyl, 2'-O,4'-C-methylene--D-ribofuranosyl (LNA) and phosphoroamidate. In general, the modifications at ribose sugar of the oligonucleotide attribute special features to the duplexes than other modifications. Molecular dynamics (MD) simulations are valuable tools to study their dynamics, conformational preferences, stabilities etc. Here, we have employed different simulation techniques such as classical MD, replica exchange MD to understand the impact of modification on the duplex properties. The simulations revealed several important aspects related to duplex stability and relation with their base content. Our results also suggest that the differential changes observed in nucleic acids depends on the position, nature of the modified site. This study also provides a molecular level picture for the origin of structural changes arise due to different types of modifications which is very important in designing the most fruitful antisense oligonucleotides.
Impact of 2'-hydroxyl sampling on the conformational properties of RNA: Update of the CHARMM all-atom additive force field for RNA
Elizabeth J. Denning,Deva Priyakumar U,Lennart Nilsson,Alexander D. Mackerell Jr
Journal of Computational Chemistry, JCC, 2011
@inproceedings{bib_Impa_2011, AUTHOR = {Denning, Elizabeth J. and U, Deva Priyakumar and Nilsson, Lennart and Jr, Alexander D. Mackerell }, TITLE = {Impact of 2'-hydroxyl sampling on the conformational properties of RNA: Update of the CHARMM all-atom additive force field for RNA}, BOOKTITLE = {Journal of Computational Chemistry}. YEAR = {2011}}
Here, we present an update of the CHARMM27 all-atom additive force field for nucleic acids that improves the treatment of RNA molecules. The original CHARMM27 force field parameters exhibit enhanced Watson-Crick base pair opening which is not consistent with experiment, whereas analysis of molecular dynamics (MD) simulations show the 2'-hydroxyl moiety to almost exclusively sample the O3' orientation. Quantum mechanical (QM) studies of RNA related model compounds indicate the energy minimum associated with the O3' orientation to be too favorable, consistent with the MD results. Optimization of the dihedral parameters dictating the energy of the 2'-hydroxyl proton targeting the QM data yielded several parameter sets, which sample both the base and O3' orientations of the 2'-hydroxyl to varying degrees. Selection of the final dihedral parameters was based on reproduction of hydration behavior as related to a survey of crystallographic data and better agreement with experimental NMR Jcoupling values. Application of the model, designated CHARMM36, to a collection of canonical and noncanonical RNA molecules reveals overall improved agreement with a range of experimental observables as compared to CHARMM27. The results also indicate the sensitivity of the conformational heterogeneity of RNA to the orientation of the 2'-hydroxyl moiety and support a model whereby the 2'- hydroxyl can enhance the probability of conformational transitions in RNA.
Characterization of ERK Docking Domain Inhibitors that Induce Apoptosis by Targeting Rsk-1 and Caspase-9
Sarice R Boston,Rahul Deshmukh,Scott Strome,Deva Priyakumar U, Alexander D MacKerell Jr,Paul Shapiro
BMC Cancer, BMCC, 2011
@inproceedings{bib_Char_2011, AUTHOR = {Boston, Sarice R and Deshmukh, Rahul and Strome, Scott and U, Deva Priyakumar and Jr, Alexander D MacKerell and Shapiro, Paul }, TITLE = {Characterization of ERK Docking Domain Inhibitors that Induce Apoptosis by Targeting Rsk-1 and Caspase-9}, BOOKTITLE = {BMC Cancer}. YEAR = {2011}}
Background: The extracellular signal-regulated kinase-1 and 2 (ERK1/2) proteins play an important role in cancer cell proliferation and survival. ERK1/2 proteins also are important for normal cell functions. Thus, anti-cancer therapies that block all ERK1/2 signaling may result in undesirable toxicity to normal cells. As an alternative, we have used computational and biological approaches to identify low-molecular weight compounds that have the potential to interact with unique ERK1/2 docking sites and selectively inhibit interactions with substrates involved in promoting cell proliferation. Methods: Colony formation and water soluble tetrazolium salt (WST) assays were used to determine the effects of test compounds on cell proliferation. Changes in phosphorylation and protein expression in response to test compound treatment were examined by immunoblotting and in vitro kinase assays. Apoptosis was determined with immunoblotting and caspase activity assays. Results: In silico modeling was used to identify compounds that were structurally similar to a previously identified parent compound, called 76. From this screen, several compounds, termed 76.2, 76.3, and 76.4 sharing a common thiazolidinedione core with an aminoethyl side group, inhibited proliferation and induced apoptosis of HeLa cells. However, the active compounds were less effective in inhibiting proliferation or inducing apoptosis in non-transformed epithelial cells. Induction of HeLa cell apoptosis appeared to be through intrinsic mechanisms involving caspase-9 activation and decreased phosphorylation of the pro-apoptotic Bad protein. Cell-based and in vitro kinase assays indicated that compounds 76.3 and 76.4 directly inhibited ERK-mediated phosphorylation of caspase-9 and the p90Rsk-1 kinase, which phosphorylates and inhibits Bad, more effectively than the parent compound 76. Further examination of the test compound’s mechanism of action showed little effects on related MAP kinases or other cell survival proteins. Conclusion: These findings support the identification of a class of ERK-targeted molecules that can induce apoptosis in transformed cells by inhibiting ERK-mediated phosphorylation and inactivation of pro-apoptoticproteins.
Inherent Conformational Preferences of the Reactants Control Their Transannular Diels-Alder Reactivity!
VANGARA PRATHYUSHA,Ramakrishna Vedantam .S,Deva Priyakumar U
Applied Theory On Molecular Systems, ATOMS, 2011
@inproceedings{bib_Inhe_2011, AUTHOR = {PRATHYUSHA, VANGARA and .S, Ramakrishna Vedantam and U, Deva Priyakumar }, TITLE = {Inherent Conformational Preferences of the Reactants Control Their Transannular Diels-Alder Reactivity!}, BOOKTITLE = {Applied Theory On Molecular Systems}. YEAR = {2011}}
Transannular Diels Alder reaction (TADA), a combination of both intramolecular and Diels-Alder reactions, is a powerful synthetic organic reaction mechanism used in synthesizing polycyclic compounds with high degree of chemo-, regio- and stereospecificity. TADA reactions occur in (x+y+2)-membered triene macrocycles, which contain both the diene and the dienophile moieties, to form A.B.C[x+6+y] type tricyclic compounds. The TADA reactivity of a series of 14-membered macrocyclic rings have been studied using the density functional B3LYP level of theory. The reactants and the transition states are capable of existing in different conformations. The key conformers of the transition states were studied at the B3LYP level. We have done force field parameters for the trienes, and performed detailed replica exchange molecular dynamics simulations to identify all possible isomers of the reactants. We established a linear correlation between the activation energies and the extent of sampling of the reactants in certain conformational states. In this presentation, we also will present the effect of the lengths of the linkers (3 to 5 methylene units) connecting the diene and dienophile units on the TADA reactivities.
Molecular Dynamics Simulations Reveal Substrate Recognition Mechanism of Ribonuclease H
Suresh Gorle,Deva Priyakumar U
Applied Theory On Molecular Systems, ATOMS, 2011
@inproceedings{bib_Mole_2011, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {Molecular Dynamics Simulations Reveal Substrate Recognition Mechanism of Ribonuclease H}, BOOKTITLE = {Applied Theory On Molecular Systems}. YEAR = {2011}}
Ribonucleases act as biological counter weights by playing a major role in gene expression. Ribonuclease H is one of them which performs the endonucleotic cleaveage reaction of the 3'-O-P linkage of RNA strand of DNA/RNA hybrids. These enzymes show specificity towards DNA/RNA hybrids but not pure structures. One of the reason for this differentiation could be mixed A/Bconformation of hybrid compared to pure structures and there might be other reasons. There is a debate concerning the substrate recognition mechanism whether this is because of induced fit type mechanism or conformational rearrangement of monomers in order to come close or may be non bonded interactions which will put the enzyme and substrate together in order to perform the hydrolysis reaction. While the role of metal ions in the reaction has been recognised, the mechanism is not obvious. For differentating the structural changes, we performed long molecular dynamics (MD) simulations on ribonuclease H enzyme of B.halodurans with its substrate and their monomers. The calculations show that the large conformational changes in both the monomers favours the reaction. Since the protein hydrophobic core is very stable, these changes will take place at surface level. The principal component analysis also supports the large conformational changes in protein and hybrid. Apart form these changes, the protein binding increases the plasticity of the hybrids. Overall the calculations show that structural deformations in both hybrid and enzyme brings them close to each other. Free energy calculations have been performed to study the proteinhybrid interactions
Effect of Thermal Perturbations and chemical denaturants on the Folding of Trp-cage miniprotein via Free Energy Calculations
KASAVAJHALA KOUSHIK,Deva Priyakumar U
Applied Theory On Molecular Systems, ATOMS, 2011
@inproceedings{bib_Effe_2011, AUTHOR = {KOUSHIK, KASAVAJHALA and U, Deva Priyakumar }, TITLE = {Effect of Thermal Perturbations and chemical denaturants on the Folding of Trp-cage miniprotein via Free Energy Calculations}, BOOKTITLE = {Applied Theory On Molecular Systems}. YEAR = {2011}}
Understanding protein folding thermodynamics and kinetics is a central issue in molecular biology. Molecular dynamics (MD) simulations are being extensively used for this purpose. However, direct comparison between simulations and experiments requires both, an accurate description of the system and extensive sampling of the conformational space. Variants of MD have been developed to overcome sampling issues. Here, we use umbrella sampling technique to quantitatively compute the differences in thermodynamic quantities of the unfolding process of Trp-cage miniprotein TC5b (PDB ID: 1L2Y) due to changes in temperature as well as the presence/absence of chemical denaturants. Urea at a concentration of 8M has been used as the chemical denaturant. WHAM was used to obtain the unbiased free energy profiles. The free energy profile can be categorised into three parts – the opening of the Trp-cage, separation of secondary structure elements and distortion of alpha-helix. An increase in temperature or presence of urea effects only the opening of the Trp-cage while the separation of the secondary structure elements and distortion of alpha-helix is independent of temperature and urea. Contour plots were constructed to substantiate the results obtained from free energy profiles.
Structural and Energetic Determinants of Thermal Stability and Hierarchical Unfolding Pathways of Hyperthermophilic Proteins, Sac7d and Sso7d
Deva Priyakumar U,Saini Shiva Rama Krishna,NAGARJUNA K R,S KARUNAKAR REDDY
The Journal of Physical Chemistry B, JPCB, 2010
@inproceedings{bib_Stru_2010, AUTHOR = {U, Deva Priyakumar and Krishna, Saini Shiva Rama and R, NAGARJUNA K and REDDY, S KARUNAKAR }, TITLE = {Structural and Energetic Determinants of Thermal Stability and Hierarchical Unfolding Pathways of Hyperthermophilic Proteins, Sac7d and Sso7d}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2010}}
Identification of the structural and energetic determinants responsible for enhancing the stability of proteins is crucial. Hyperthermophilic proteins are naturally occurring proteins that exhibit high thermal stability and are good candidates for the investigation and understanding of structure-stability relationships. Sac7d from Sulfolobus acidocaldarius and Sso7d from Sulfolobus solfactaricus are two homologous hyperthermophilic proteins that were shown to be quite stable at high temperatures. Molecular dynamics simulations at the nanosecond time scale at different temperatures were performed to examine the factors affecting their stability. The three-dimensional structures of these proteins were observed to be similar to the experimental structure at 300 and 360 K but were found to undergo denaturation at 500 K. Both proteins exhibit similar unfolding pathways that correlates well with the calculated intermolecular interaction energies. The differential dynamic behaviors of these molecules at different temperatures were examined. Structural and energetic analysis of the contributions of salt bridges indicates a stabilizing effect at higher temperatures. However, the lifetimes of the salt bridges were found to be quite short, and several new salt bridges formed at 500 K supporting previous studies that the desolvation penalty due to the formation of salt bridges decreases at elevated temperatures. Hydrophobic interactions, which decrease with increase in temperature, were also found to be crucial in the stability of these proteins. Overall, the study shows that a balance among the salt bridge interactions, hydrophobic interactions, and solvent properties is primarily responsible for the high thermal stability of this class of proteins.
Atomistic Details of the Ligand Discrimination Mechanism of SMK/SAM-III Riboswitch
Deva Priyakumar U
The Journal of Physical Chemistry B, JPCB, 2010
@inproceedings{bib_Atom_2010, AUTHOR = {U, Deva Priyakumar }, TITLE = {Atomistic Details of the Ligand Discrimination Mechanism of SMK/SAM-III Riboswitch}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2010}}
SAM-III riboswitch, involved in regulating sulfur metabolic pathways in lactic acid bacteria, is capable of differentiating S-adenosyl-L-methionine (SAM) from its structurally similar analog S-adenosyl-L-homocysteine (SAH). Atomic level understanding of the ligand recognition mechanism of riboswitches is essential for understanding their structure-function relationships in general. In the present study, we have employed molecular dynamics (MD) simulations on five model systems to elucidate the discrimination mechanism adopted by the SAM-III riboswitch that enables differential binding of SAM with respect to SAH. The structures of the binding pocket of the riboswitch, and the modes of binding of the adenine moiety of SAM obtained from the MD simulations are similar to the experimental structure. However, MD simulations of the riboswitch-SAH complexes lead to partial unbinding of the ligand and structural changes in the RNA binding pocket. Detailed analyses were performed to examine the structural and energetic factors involved in such a differentiation. The calculations reveal a novel mechanism by which the aptamer domain specifically recognizes the adenine moiety of SAM/SAH, but SAM is better stabilized in the binding pocket due to nonspecific electrostatic interactions involving the sulfonium group. Additionally, the results support less dependence of the ligand conformation in the bound form on the effective binding of SAM to the riboswitch.
Sequence Effects on the Structures and Dynamics of DNA-RNA Hybrids
Suresh Gorle,Deva Priyakumar U
National Symposium on ‘Recent Trends in BioPhysics, NSRTB, 2010
@inproceedings{bib_Sequ_2010, AUTHOR = {Gorle, Suresh and U, Deva Priyakumar }, TITLE = {Sequence Effects on the Structures and Dynamics of DNA-RNA Hybrids}, BOOKTITLE = {National Symposium on ‘Recent Trends in BioPhysics}. YEAR = {2010}}
DNA-RNA hybrids are heterogeneous nucleic acid duplexes that contain a DNA strand and a RNA strand. They are identified as key intermediates in transcription, and in other biological processes. These novel molecules have also been recognized as potential candidates in antisense therapy. While DNA exists in B-form, and RNA in A-form, the process of hybridization between these two distinct conformations, and the resultant structural features of the hybrid duplexes are interesting in their own right. It has been suggested that hybrids have unique conformational properties that form the basis of Ribonuclease H (RNase H) activity only on these molecules, and not on pure DNA or RNA. Quantitative examination of the conformational properties of hybrids is very important for understanding the RNase H activity. Several nanosecond long molecular dynamics (MD) simulations were performed on all the available structures in presence of explicit solvent environment. All the MD simulations were performed using the CHARMM27 all atom nucleic acid force field. In addition to the MD simulations on hybrids, control simulations were also performed on pure DNA and RNA duplexes with identical sequences. Several structural, dynamic and energetic properties including deviations, flexibility, intramolecular entropy, helical properties, and backbone conformational preferences were computed. The conformational properties of the sugar puckering angles revealed interesting trends that depend on the sequence of the duplex. RNA strands in all the duplexes sample conformational regions corresponding to a pure-A form nucleic acid. However, the DNA strands adopt mixing of A- and B- type conformations resulting in intermediate structures for the hybrids. The extent of this mixing depends primarily on the relative purine to pyrimidine composition ratio in the DNA strands, and sequence effects. The presentation will focus on the structural and dynamic properties of hybrid duplexes in comparison with those of pure-RNA and DNA duplexes.
Role of the Adenine Ligand on the Stabilization of the Secondary and Tertiary Interactions in the Adenine Riboswitch
Deva Priyakumar U,Alexander D. Mackerell
Journal of Molecular Biology, JMB, 2010
@inproceedings{bib_Role_2010, AUTHOR = {U, Deva Priyakumar and Mackerell, Alexander D. }, TITLE = {Role of the Adenine Ligand on the Stabilization of the Secondary and Tertiary Interactions in the Adenine Riboswitch}, BOOKTITLE = {Journal of Molecular Biology}. YEAR = {2010}}
Riboswitches are RNA-based genetic control elements that function via a conformational transition mechanism when a specific target molecule binds to its binding pocket. To facilitate an atomic detail interpretation of experimental investigations on the role of the adenine ligand on the conformational properties and kinetics of folding of the add adenine riboswitch, we performed molecular dynamics simulations in both the presence and the absence of the ligand. In the absence of ligand, structural deviations were observed in the J23 junction and the P1 stem. Destabilization of the P1 stem in the absence of ligand involves the loss of direct stabilizing interactions with the ligand, with additional contributions from the J23 junction region. The J23 junction of the riboswitch is found to be more flexible, and the tertiary contacts among the junction regions are altered in the absence of the adenine ligand; results suggest that the adenine ligand associates and dissociates from the riboswitch in the vicinity of J23. Good agreement was obtained with the experimental data with the results indicating dynamic behavior of the adenine ligand on the nanosecond time scale to be associated with the dynamic behavior of hydrogen bonding with the riboswitch. Results also predict that direct interactions of the adenine ligand with U74 of the riboswitch are not essential for stable binding although it is crucial for its recognition. The possibility of methodological artifacts and force-field inaccuracies impacting the present observations was checked by additional molecular dynamics simulations in the presence of 2,6-diaminopurine and in the crystal environment. Copyright (c) 2010 Elsevier Ltd. All rights reserved.
Molecular simulations on the thermal stabilization of DNA by hyperthermophilic chromatin protein Sac7d, and associated conformational transitions
Deva Priyakumar U,Harika.G,Suresh Gorle
The Journal of Physical Chemistry B, JPCB, 2010
@inproceedings{bib_Mole_2010, AUTHOR = {U, Deva Priyakumar and Harika.G, and Gorle, Suresh }, TITLE = {Molecular simulations on the thermal stabilization of DNA by hyperthermophilic chromatin protein Sac7d, and associated conformational transitions}, BOOKTITLE = {The Journal of Physical Chemistry B}. YEAR = {2010}}
Sac7d belongs to a family of chromosomal proteins, which are crucial for thermal stabilization of DNA at higher growth temperatures. It is capable of binding DNA nonspecifically, and is responsible for the increase in the melting temperature of DNA in the bound form up to 85 °C. Molecular dynamics (MD) simulations were performed at different temperatures on two protein-DNA complexes of Sac7d. Various structural and energetic parameters were calculated to examine the DNA stability and to investigate the conformational changes in DNA and the protein-DNA interactions. Room temperature simulations indicated very good agreement with the experimental structures. The protein structure is nearly unchanged at both 300 and 360 K, and only up to five base pairs of the DNA are stabilized by Sac7d at 360 K. However, the MD simulations on DNA alone systems show that they lose their helical structures at 360 K further supporting the role of Sac7d in stabilizing the oligomers. At higher temperatures (420 and 480 K), DNA undergoes denaturation in the presence and the absence of the protein. The DNA molecules were found to undergo B- to A-form transitions consistent with experimental studies, and the extent of these transitions are examined in detail. The extent of sampling B- and A-form regions was found to show temperature and sequence dependence. Multiple MD simulations yielded similar results validating the proposed model. Interaction energy calculations corresponding to protein-DNA binding indicates major contribution due to DNA backbone, explaining the nonspecific interactions of Sac7d.
Computational studies on the transannular Diels-Alder reactions of macrocyclic trienes
VANGARA PRATHYUSHA,K RAMAKRISHNA SYAMKISHOR,Deva Priyakumar U
Theoritical Chemistry Symposium, TCS, 2010
@inproceedings{bib_Comp_2010, AUTHOR = {PRATHYUSHA, VANGARA and SYAMKISHOR, K RAMAKRISHNA and U, Deva Priyakumar }, TITLE = {Computational studies on the transannular Diels-Alder reactions of macrocyclic trienes}, BOOKTITLE = {Theoritical Chemistry Symposium}. YEAR = {2010}}
One of the most efficient chemical transformations in the natural product synthesis of steroid like molecules is the synthetic methodology that combines the pericyclic and intramolecular reactions termed as transannular Diels-Alder reaction (TADA). TADA reactions occur in (x+y+2)-membered triene macrocycles, which contain both the diene and the dienophile, to form A.B.C [x.6.y] type tricyclic compounds. Initially, a validation study was carried out to identify an appropriate method for studying this reaction pathway. Several methods including HF, post-HF methods such as Mpx and CC, and density functional B3LYP level were considered, and we found that B3LYP method used along with a reasonably smaller basis set (6-31G(d)) is adequate enough. We have extensively studied the TADA reactions of six possible cis/trans-isomers of 14-membered macrocycles, and correlated their reactivities with experimental data wherever applicable. We also have studied the effect of ring size on the TADA reactivity by performing calculations on 552, 562, 572, 652, 672, 752, 762 and 772 systems using the B3LYP level employing the 6-31G(d) and the cc-pVTZ basis sets. We have shown that the activation energy and the reaction energy decrease with respect to the increase in the ring size. In addition to the activation energies, the conformational preferences of the reactants were suggestd to be crucial for the reaction to take place, and on its stereochemical outcome. We have studied their conformational properties using replica exchange molecular dynamics simulations. Initially, we obtained force field parameters for the macrocyclic trienes based on MP2 calculations on six model systems. The analysis concentrated on the dihedral angle corresponding to the central bond of the diene moiety, and the distance between the reaction centers. The extent of conformational sampling of each of the macrocycles is compared with the reactivities. The presentation will discuss the TADA reactivity in terms of activation energies corresponding to the transition states of the reactions, and also based on the conformational properties of the reactants.
Role of nonbonded interactions on the thermal stability of biomolecules in thermophiles
BIKKINA SWETHA,Deva Priyakumar U
Theoritical Chemistry Symposium, TCS, 2010
@inproceedings{bib_Role_2010, AUTHOR = {SWETHA, BIKKINA and U, Deva Priyakumar }, TITLE = {Role of nonbonded interactions on the thermal stability of biomolecules in thermophiles}, BOOKTITLE = {Theoritical Chemistry Symposium}. YEAR = {2010}}
Hyperthermophilic proteins are naturally occurring proteins that are capable of retaining their structure, and performing their normal functions at temperatures above 80º C. Identification of the structural determinants that are crucial for such a remarkable stability is of fundamental importance, and has potential applications in biocatalysis. Molecular dynamics simulations on two such proteins (Sac7d and Sso7d) illustrated the importance of ion pair interactions for achieving thermostability.1 Previous studies have indicated that the formation of ion pair interactions does not contribute to stability due to high desolvation penalty, while others have argued in favor of stabilization. We have performed umbrella sampling simulations on model systems at different temperatures, which showed that with increasing temperature, the stability of the ion pair interactions become highly favorable. Thus, protein folding involving association of an ion pair may lead to stabilization at enhanced temperature close to the growth temperature of the hyperthemophiles. We have also investigated the effect of protein-DNA interactions, and their role in DNA stabilization at these conditions.2 Canonical B- to A-form transitions were observed in the protein-DNA complexes, and the protein-DNA interaction energies are used to explain the sequence nonspecific nature of the binding activity. The crucial nonbonded interactions that facilitate extraordinary thermal stability of proteins and DNAs in thermophiles will be discussed in this presentation.
The interactions of HIV accessory proteins and host proteins as targets for antiviral drug development
Rahul Arya, Akankshi Munjal,Madeeha Aqil,Nabab Khan,GANGAMPALLI HARIKA,Satyajit Mayor, Satyajit Rath,Deva Priyakumar U,Shahid Jameel
International Conference on Antivirals for Neglected and Emerging Viruses, ICAV, 2010
@inproceedings{bib_The__2010, AUTHOR = {Arya, Rahul and Munjal, Akankshi and Aqil, Madeeha and Khan, Nabab and HARIKA, GANGAMPALLI and Mayor, Satyajit and Rath, Satyajit and U, Deva Priyakumar and Jameel, Shahid }, TITLE = {The interactions of HIV accessory proteins and host proteins as targets for antiviral drug development}, BOOKTITLE = {International Conference on Antivirals for Neglected and Emerging Viruses}. YEAR = {2010}}
The enlarging HIV/AIDS pandemic and the virus’ ability to develop resistance to anti-retroviral therapy requires the development of new drugs based on novel targets. Viruses interact extensively with the host at the cellular and molecular levels, which is necessary for the expression of viral proteins and replication of viral genomes. At another level, these interactions also help the virus evade host immune responses. Besides the prototypic retroviral proteins (Gag, Pol and Env), HIV-1 also expresses two regulatory proteins (Tat and Rev) and four accessory proteins (Nef, Vif, Vpr and Vpu). The accessory proteins are not required for viral replication in vitro, but are critical for pathogenesis and disease development in the host. We have focused our efforts on two of these – Nef and Vpu, with an aim to understand their interactions with host proteins and the functional consequences of these interactions for HIV-1 infection and pathogenesis. The Nef protein binds to multiple host proteins to optimize the host cell environment and to enable infected cells to evade the host immune response. A major property of Nef is to downregulate surface expression of the CD4, MHC I and MHC II proteins, by increasing the endocytosis of these molecules. We also discovered that Nef binds to the cytoplasmic domains of the B7 family of costimulatory proteins – CD80 and CD86; this results in their removal from the cell surface and sequestration to the Golgi compartment (1-3). We have now developed an in vitro screen based on purified recombinant Nef proteins and a 20-mer synthetic peptide from the cytoplasmic domain of CD80 and CD86. A chemical library comprising of 1064 compounds was screened and ~7% compounds were obtained as hits, which inhibited the Nef-CD80/CD86 interaction. These compounds are now being evaluated in a cellbased assay and an independent dose dependent response assay. The top hits will be used in a combinatorial chemical approach to synthesize a large array of new molecules for further screening using the in vitro and cell-based screens. The HIV-1 Vpu protein is required for the release of new virions from the surface of infected cells; for this the transmembrane domain of Vpu is critical. Using the Topology Data Bank of Transmembrane Proteins (TOPDB) database we extracted 13 alpha-helical sequences that closely resemble the Vpu-TM domain. We have taken the top two hits and have constructed recombinant Vpu proteins in which the native TM domain was replaced by the identified sequences. These recombinant proteins were successfully expressed in HEK293 cells, and are now being tested for their effects on virion release as well as their ability to inhibit Vpu activity. We describe here two paradigms to screen for anti-HIV activity based on proteinprotein interactions. Results will be presented for both of these.
Effect of thermal perturbation, and chemical denaturant on the mechanical unfolding of proteins via free energy calculations and steered molecular dynamics simulations
NAVNEET BUNG,KASAVAJHALA KOUSHIK,Deva Priyakumar U
Theoritical Chemistry Symposium, TCS, 2010
@inproceedings{bib_Effe_2010, AUTHOR = {BUNG, NAVNEET and KOUSHIK, KASAVAJHALA and U, Deva Priyakumar }, TITLE = {Effect of thermal perturbation, and chemical denaturant on the mechanical unfolding of proteins via free energy calculations and steered molecular dynamics simulations}, BOOKTITLE = {Theoritical Chemistry Symposium}. YEAR = {2010}}
Thermal, chemical and force induced unfolding are three primary experimental approaches that have been successfully used to study protein folding and associated pathways. Recent experimental studies have demonstrated the importance of the use of combination of these techniques to study various aspects of protein folding. We have addressed this issue using two different model systems. The effect of temperature on the mechanical unfolding of titin I27 immunoglobulin domain has been investigated using steered molecular dynamics (MD) simulations performed both at constant velocity and constant force. The maximum force of unfolding decreases with respect to the increase in the temperature, but the extension corresponding to this force remains similar. However, we observed that the maximum probable unfolding force steadily decreases with an increase in temperature (see below). Several computations done on each of the temperatures consistently reproduce these effects. The results will be discussed in light of the available experimental data, and molecular level details on the conformational changes that effect such temperature dependence will be presented. In another study, we have done rigorous free energy calculations on the unfolding of Trp-mini cage protein using umbrella sampling MD simulations. Calculations were done with respect to increase in the temperature, and with respect to the concentration of the urea solution. The change in the free energy profiles, and the effect of external perturbations, and the atomistic details of the unfolding will be discussed. Temperature
Quantum chemical study of control of stereochemistry of tetrahdyrofuran ring in tripeptides containing tetrahydrofuran amino acids
Suresh Kumar N V,Deva Priyakumar U,Harjinder Singh,T. K. Chakraborty
International Conference on Physics Biology Interface, ICPBI, 2009
@inproceedings{bib_Quan_2009, AUTHOR = {V, Suresh Kumar N and U, Deva Priyakumar and Singh, Harjinder and Chakraborty, T. K. }, TITLE = {Quantum chemical study of control of stereochemistry of tetrahdyrofuran ring in tripeptides containing tetrahydrofuran amino acids}, BOOKTITLE = {International Conference on Physics Biology Interface}. YEAR = {2009}}
Design of biomimetic systems is an alternative method of finding biologically active molecules. Peptides incorporating tetrahydrofuran amino acids (TAA) adopt secondary folding patterns and thus mimic natural peptides. The preferred conformation of TAA derived linear and cyclized peptides depends on the stereochemistry of the tetrahydrofuran (THF) ring. Control of stereochemistry of THF ring in the structure of the linear tripeptide Boc-TAA-Leu-Val- OMe containing (2R,5S)-cis TAA, (2S,5R)-cis TAA, (2R,5R)-trans TAA and (2S,5S)-trans TAA is investigated using structural parameters and energetics data obtained from density functional theory (DFT) based calculations at B3LYP/6-31G(d,p) level of theory. We found that the conformations associated with -, -turn structures are stabilized by intramolecular hydrogen bonding interactions. Kinetic control of reactions leading to intra and inter molecular cyclization of tripeptide TAA-Gly-Gly containing (2S,5R)-cis TAA and (2S,5S)-trans TAA is studied at the same level of theory in gas phase.[1] We observe that kinetic control favors cyclodimerization instead of intramolecular cyclization.
Conformational Analysis of Macrocyclic Trienes: Force Field Optimization and Molecular Dynamics Studies
S. Ramakrishna,Deva Priyakumar U
Of Molecules and Materials, OMAM, 2009
@inproceedings{bib_Conf_2009, AUTHOR = {Ramakrishna, S. and U, Deva Priyakumar }, TITLE = {Conformational Analysis of Macrocyclic Trienes: Force Field Optimization and Molecular Dynamics Studies}, BOOKTITLE = {Of Molecules and Materials}. YEAR = {2009}}
Macrocyclic trienes are important precursors for the synthesis of several biologically relevant molecules such as steroids. These molecules undergo (4+2) cycloaddition to form tricyclic compounds, which has four newly formed stereocenters. The stereochemistry at these four carbon centers has been shown to be largely controlled by the reaction energy barriers, and the conformational preferences of the macrocycle. Molecular dynamics (MD) simulation is a highly efficient approach for studying conformational properties of chemical molecules and biological macromolecules. MP2/6-31G* method was used to obtain potential energy surfaces corresponding to appropriate model compounds and used as the target data. The force field parameters were refined to accurately represent the target data involving all the model compounds chosen. Using the refined molecular mechanics force field, MD simulations were performed on six possible 14-membered symmetric trienes at different temperatures. We have also used replica exchange MD (RexMD) simulations with eight replicas between 300 and 370 K. The molecules were found to sample a larger conformational space in the RexMD simulation compared to the regular MD simulations. The probability distributions corresponding to the dihedral angles, and distances between the reaction centers have been analyzed. The relationship between the conformational sampling of the macrocyclic trienes vis-à-vis their reactivity will be discussed. Additionally, clustering of the structures from the MD trajectories was done to identify all unique conformations using a new approach.
Detailed Investigation of the Thermal Stabilization of DNA by Sac7d
G Harika,Deva Priyakumar U
Of Molecules and Materials, OMAM, 2009
@inproceedings{bib_Deta_2009, AUTHOR = {Harika, G and U, Deva Priyakumar }, TITLE = {Detailed Investigation of the Thermal Stabilization of DNA by Sac7d}, BOOKTITLE = {Of Molecules and Materials}. YEAR = {2009}}
Sac7d is a protein present in Sulfolobus acidocaldarius, a hyperthermophilic archaeon, which is stable at extreme conditions. It is a chromosomal protein, which is capable of binding DNA thereby increasing its melting temperature. It was suggested to stabilize central 3 to 5 base pair steps in DNA. It binds to the minor groove of the DNA causing a kink in the helix without any sequential preference. We have used molecular dynamics (MD) simulations to study the binary complexes of Sac7d bound to DNA at 300 and 360 K. The objective is to understand the atomic detailed mechanism of the capability of Sac7d in facilitating DNA stability at higher temperature. As has been shown before, the structure of the protein is highly stable both at 300 and 360 K. The DNA duplex was found to be stable at 300 K and is similar to the X-ray crystal structure. Consistent with experiments, at least the central four base pairs of the DNA were found to be base paired during most of the simulation at 360 K. The dynamic properties of the protein in the protein-DNA complexes remained unchanged compared to when the protein is unbound. Interestingly, B- to A-form transitions occur in both the DNA molecules when bound to Sac7d, which was confirmed by sugar puckering angles, and backbone dihedral angles. Detailed examination of the protein-DNA interactions reveals that the stabilization of the DNA molecules is mostly due to the favorable interaction of the protein with the backbone of the DNA.
Investigation of Base Flipping in DNA bound to HhaI Methyltransferase and DNA in Solution via Free Energy Calculations
Deva Priyakumar U,Alexander D. Mackerell
National Symposium on Cellular and Molecular Biophysics, NCMB, 2009
@inproceedings{bib_Inve_2009, AUTHOR = {U, Deva Priyakumar and Mackerell, Alexander D. }, TITLE = {Investigation of Base Flipping in DNA bound to HhaI Methyltransferase and DNA in Solution via Free Energy Calculations}, BOOKTITLE = {National Symposium on Cellular and Molecular Biophysics}. YEAR = {2009}}
Base flipping is a process by which one of the bases of the DNA swings out from its helical position to an extrahelical form. Several proteins such as methyltransferases, glycosylases and endonucleases employ this phenomenon in order to gain access to the target base for chemical modification. Even in the absence of protein, base pair opening occurs in nucleic acids in solution. X-ray crystal studies, fluorescence spectroscopy and NMR imino proton exchange experiments have been extensively employed to study the structures and dynamics involved in this fascinating event. We have employed molecular dynamics simulations to study base flipping both in presence and absence of the protein. The free energy pathways obtained using potential of mean force calculations will be presented. The base pair opening dynamics of AT and GC base pairs was investigated by calculating the equilibrium constants corresponding to the equilibrium between the base pair open and closed states, and compared with experimental data. Our calculations showed that the contribution to the overall equilibrium constants of the base pair opening is primarily from the purine bases. The methodological issues such as sufficient sampling and force field parameters will be addressed. In another study, the role of THR250 HhaI methyltransferase in base flipping will be discussed. THR250 is a highly conserved residue in methyltransferases and was proposed to be crucial for the correct placement of flipped cytosine for methylation. Previous computational studies have shown that the protein actively facilitates the base flipping process. Free energy calculations of base flipping in the presence of the wild type and T250G, T250A and T250D mutants of the protein were computed. The free energy profiles, structural changes and the energetic parameters for the base flipping process will be presented.
Denaturation of RNA by Urea: A Molecular Dynamics Study
Deva Priyakumar U,D. Thirumalai,Alexander D. MacKerell Jr
Theoritical Chemistry Symposium, TCS, 2009
@inproceedings{bib_Dena_2009, AUTHOR = {U, Deva Priyakumar and Thirumalai, D. and Jr, Alexander D. MacKerell }, TITLE = {Denaturation of RNA by Urea: A Molecular Dynamics Study}, BOOKTITLE = {Theoritical Chemistry Symposium}. YEAR = {2009}}
Urea concentration dependent denaturation of proteins and nucleic acids has been widely used in obtaining (un)folding pathways and in calculating the free energies. Molecular dynamics simulations were performed on a 22-nt RNA hairpin in aqueous solution environment and in different concentrations (1-8M) of urea. The energetic and structural aspects of the role of urea in denaturating the RNA hairpin were investigated. In agreement with experimental studies, the RNA undergoes denaturation at higher concentrations of urea (> 6M). Surprisingly, the hairpin was observed to be more stable at lower concentrations of urea (1-2M) even compared to the RNA in aqueous environment. Various structural, energetic and solvation properties were analyzed to understand the mechanism of the interaction of urea with the hairpin. It was observed that urea does not directly facilitate the opening of the base pairs that leads to denaturation at higher concentrations. Instead, the open states of the RNA are stabilized by via hydrogen bonding interactions and stacking interactions thereby stabilizing the denatured form of the RNA.
Quantum Mechanical Studies on the Transannular Diels-Alder Reactions, and Related Pericyclic Reactions of 14-Membered Macrocyclic Trienes
VANGARA PRATHYUSHA,Deva Priyakumar U
Of Molecules and Materials, OMAM, 2009
@inproceedings{bib_Quan_2009, AUTHOR = {PRATHYUSHA, VANGARA and U, Deva Priyakumar }, TITLE = {Quantum Mechanical Studies on the Transannular Diels-Alder Reactions, and Related Pericyclic Reactions of 14-Membered Macrocyclic Trienes}, BOOKTITLE = {Of Molecules and Materials}. YEAR = {2009}}
Transannular Diels-Alder reaction (TADA) combines traditional pericyclic and intramolecular reactions, and has been shown to be one of the most efficient chemical transformations for synthesizing complicated natural products. TADA reactions occur in (x+y+2)-membered triene macrocycles, which contain both the diene and the dienophile, to form A.B.C[x.6.y] type tricyclic compounds. In addition, these macrocycles were shown to interconvert among its isomers via [1,5]-sigmatropic hydrogen shifts. The stereochemical outcome of the TADA reactions depends on the energetics of various possible transition states. Initially, a validation study was undertaken to assess the performance of various ab initio (HF, MPx, and CC) and hybrid density functional B3LYP methods using a variety of basis sets. Comparison of the reaction barriers, and reaction energies from various methods indicate that B3LYP/6-31G* level is appropriate enough for accurately modeling this class of reactions. All possible reaction pathways involving the TADA and 1,5-sigmatropic reactions of six different 14-membered triene molecules were studied using B3LYP/cc-pVTZ were calculated. Based on the reaction energy barriers and conformational properties of the reactants, the reactivities of these molecules will be discussed.
Quick versus Right Answer: Protein Flexibility in Structure Based Drug Design
NAGARJUNA K R,Deva Priyakumar U
Biobytes, Bb, 2009
@inproceedings{bib_Quic_2009, AUTHOR = {R, NAGARJUNA K and U, Deva Priyakumar }, TITLE = {Quick versus Right Answer: Protein Flexibility in Structure Based Drug Design}, BOOKTITLE = {Biobytes}. YEAR = {2009}}
Proteins, which are the most common drug targets, are flexible and can exist in various conformers that are energetically competitive. Hence, it is crucial to consider its dynamic behavior. Most notable methods that incorporate flexibility of protein are discussed here. Structural knowledge of molecular targets (or drug receptors), usually proteins, has transformed the drug design process in the last three decades or so. While most of the drugs that were discovered 30 years ago are by serendipity and ‘trial and error’, rational drug design approaches are more focused. Rational drug design techniques use structure of the drug receptor or structure of ligands that are shown to bind to the target to identify potential drug candidates. Approaches that use the structure of drug receptor for drug discovery are referred to as structure based drug design (SBDD). X-ray crystallography and NMR spectroscopy have been in the forefront in protein structure determination. Several computational methods have been developed and new methods are being proposed for protein structure prediction from the primary sequence (for eg. homology modeling); however, they suffer from several caveats. Currently, use of SBDD has become a standard exercise as part of drug discovery and development both in industry and academia. Typically, the process involves obtaining the structure of the target protein; identification of active site; virtual screening of a small molecule database (containing chemically diverse structures of small molecules in the order of a million) and identification of potential ligands based on a chosen scoring function. Ideally, docking process involved in virtual screening and calculating the binding energy should allow the protein and the ligand to change their conformations to effectively bind to each other. Ligands being smaller compounds (typically containing tens of nonhydrogen atoms), their conformational changes can be modeled with reasonable accuracy. However, explicit modeling of protein flexibility involves a lot of computer time especially considering that a million compounds have to be screened and hence is not feasible. In short, a compromise between ‘quick answer’ and ‘right answer’ is made in docking calculations for practical applications. Recently, allowing protein to be flexible at least partially has become possible, thanks to development of efficient algorithms and, availability of faster processors, larger storage and RAM. Several approximations have been proposed to accommodate protein flexibility of which few of the notable ones are discussed in the article. The readers are suggested to refer additional resources on structure based drug design, which are given at the end.
Stability, and Hierarchical Unfolding Pathways of Hyperthermophilic Proteins and their DNA Complexes
Deva Priyakumar U
Modeling Chemical and Biological (Re) Activity, MCBA, 2009
@inproceedings{bib_Stab_2009, AUTHOR = {U, Deva Priyakumar }, TITLE = {Stability, and Hierarchical Unfolding Pathways of Hyperthermophilic Proteins and their DNA Complexes}, BOOKTITLE = {Modeling Chemical and Biological (Re) Activity}. YEAR = {2009}}
Proteins from hyperthermophilic microorganisms possess unique structurefunction properties for surviving high temperatures and for exhibiting optimal activity. In an attempt to understand the structural and energetic factors that are responsible for the thermal stability, molecular dynamics (MD) simulations were performed on Sso7d and Sac7d chromosomal proteins from S. solfataricus and S. acidocaldarius respectively. Consistent with experimental data, the proteins are quite stable at 300 and 360 K, but were found to undergo denaturation at higher temperature. Both proteins exhibit similar hierarchical unfolding pathways, which is explained based on the calculated intramolecular interaction energies. Differential dynamic behaviors of these proteins were examined, and possible roles of enzymatic activity of Sso7d are proposed. Structural, energetic, dynamic and solvation properties that influence the stability of these proteins will be discussed. The importance of hydrophobic interactions on their thermal stability will be presented using calculations on a mutant (F31A in Sso7d). These proteins are also capable of binding to DNA in a nonspecific fashion, and upon binding, melting temperature of the DNA was found to increase by about 40 K. MD simulations were also performed on protein-DNA complexes to investigate the stability of the oligonucleotide. Possible factors that facilitate the stability of these binary complexes in extraordinary thermal conditions will be discussed based on intra- and intermolecular interaction energies, correlated movements between the protein and DNA, solvation properties, and other structural parameters.
Molecular Dynamics Simulations – A Versatile Tool in Computer Aided Drug Design
Deva Priyakumar U
National Seminar on Molecular Modelling and Drug Design, NSMMDD, 2008
@inproceedings{bib_Mole_2008, AUTHOR = {U, Deva Priyakumar }, TITLE = {Molecular Dynamics Simulations – A Versatile Tool in Computer Aided Drug Design}, BOOKTITLE = {National Seminar on Molecular Modelling and Drug Design}. YEAR = {2008}}
Computational approaches are being successfully utilized for the design of small molecules that could be potential substrates for various biological targets both in academia and pharmaceutical industry. Over the past two decades or so, various computational techniques have been developed and improvements on the existing methodologies were proposed to accomplish high accuracy/ predictability of molecular properties. Consideration of the flexible nature of proteins in structure based drug design studies has been a bottleneck mainly due to the huge demand of computational and actual time. Hence most of the docking studies consider the biological macromolecules to be rigid and allow only the conformational changes of the small molecules. Similarly, the ligand based drug design approaches tend to focus more on the global minima of the small molecules. This could lead to incorrect model and hence false predictions since this approximation may not be true in cases where the global minima essentially is not the bioactive conformer – the conformer that binds to the target. Recently, advanced methods have been proposed to consider the flexible nature of the molecules. Molecular dynamics (MD) simulations have been shown to be a straightforward means to include the flexibility of both ligands and biomolecular targets. MD simulations are invaluable computational methods that are extensively used for deriving information of the structure, dynamics and energetic properties of various biomolecules including proteins, nucleic acids and lipids. The need for consideration of dynamic nature of the biomolecular targets will be demonstrated and the advantages of MD simulations to overcome this limitation will be discussed. Basic principles of MD simulations will be discussed and an introduction to biomolecular force fields will be given. Illustrated examples of the applications of MD simulations to ligand based and structure based drug design approaches will be presented.
Copyright © 2023 All rights reserved