Biology subjects

Jatana, N.

Publications and source records attributed to Jatana, N..

6 recordsLinked to original sources

bioRxiv2023

PandoraRLO: DQN and Graph convolution based method for optimized ligand pose

Predicting how proteins interact with small molecules is a complex and challenging task in the field of drug discovery. Two important aspects in this are shape complementarity and inter molecular interactions which are highly driven by the binding site and the ultimate pose of the ligand in which it interacts with the protein. Various state of the art methods exist which provide a range of ligand poses that are potentially a good fit for a given specific receptor, these are usually compute intensive and expensive. In this study, we have designed a method that provides a single optimized ligand pose for a specific receptor. The method is based on reinforcement learning where when exposed to a diverse protein ligand data set the agent is able to learn the underlying complex biochemistry of the protein ligand pair and provide an optimized pair. As a first study on usage of reinforcement learning for optimized ligand pose, the PandoraRLO model is able to predict pose within a range of 0.5[A] to 4[A] for a large number of test complexes. This indicates the potential of reinforcement learning in uncovering the inherent patterns of protein-ligand pair in 3D space.

bioinformatics↗

bioRxiv2022

PandoraRL: DQN and Graph Convolution based ligand pose learning for SARS-COV1 Mprotease

The ability to predict the correct ligand binding pose for proteinligand complex is vital for drug design. Recently several machine learning methods have suggested knowledge based scoring functions for binding energy prediction. In this study, we propose a reinforcement learning (RL) based model, PandoraRL, where the RL agent helps the ligand traverse to the optimal binding pose. The underlying representation of molecules utilizes generalized graph convolution to represent the protein ligand complex with various atomic and spatial features. The representation consists of edges formed on the basis of inter molecular interactions such as hydrogen bonds, hydrophobic interactions, etc, and nodes representing atomic features. This study presents our initial model which can train on a protein-ligand pair and predict optimal binding pose for a different ligand with the same protein. To the best of our knowledge, this is the first time an RL based approach has been put forward for predicting optimized ligand pose. CCS CONCEPTSO_LIComputing methodologies [->] Reinforcement learning. C_LI

bioinformatics↗

bioRxiv2021

Reinforcement Learning Based Approach for Ligand Pose Prediction

Identification of the potential binding site and the correct ligand pose are two crucial steps among the various steps in protein ligand interaction for a novel or known target. Currently most of the deep learning methods work on protein ligand pocket datasets for various predictions. In this study, we propose a reinforcement learning (RL) based method for predicting the optimized ligand pose where the RL agent also identifies the binding site based on its training. In order to apply various reinforcement learning techniques, we suggest a novel approach to represent the protein ligand complex using graph CNN which would help utilize both atomic and spatial features. To the best of our knowledge, this is the first time an RL based approach has been put forward for predicting optimized ligand pose.

bioinformatics↗

bioRxiv2021

A rigorous framework for detecting SARS-CoV-2 spike protein mutational ensemble from genomic and structural features

The recent release of SARS-CoV-2 genomic data from several countries has provided clues into the potential antigenic drift of the coronavirus population. In particular, the genomic instability observed in the spike protein necessitates immediate action and further exploration in the context of viralhost interactions. By temporally tracking 527,988 SARS-CoV-2 genomes, we identified invariant and hypervariable regions within the spike protein. We evaluated combination of mutations from SARS-CoV-2 lineages and found that maximum number of lineage-defining mutations were present in the N-terminal domain (NTD). Based on mutant 3D-structural models of known Variants of Concern (VOCs), we found that structural properties such as accessibility, secondary structural type, and intra-protein interactions at local mutation sites are greatly altered. Further, we observed significant differences between intra-protein networks of wild-type and Delta mutant, with the latter showing dense intra-protein contacts. Extensive molecular dynamics simulations of D614G mutant spike structure with hACE2 further revealed dynamic features with 47.7% of mutations mapping on flexible regions of spike protein. Thus, we propose that significant changes within spike protein structure have occurred that may impact SARS-CoV-2 pathogenesis, and repositioning of vaccine candidates is required to contain the spread of COVID-19 pathogen.

genomics↗

bioRxiv2020

Integrated genomic view of SARS-CoV-2 in India

India first detected SARS-CoV-2, causal agent of COVID-19 in late January-2020, imported from Wuhan, China. March-2020 onwards; importation of cases from rest of the countries followed by seeding of local transmission triggered further outbreaks in India. We used ARTIC protocol based tiling amplicon sequencing of SARS-CoV-2 (n=104) from different states of India using a combination of MinION and MinIT from Oxford Nanopore Technology to understand introduction and local transmission. The analyses revealed multiple introductions of SARS-CoV-2 from Europe and Asia following local transmission. The most prevalent genomes with patterns of variance (confined in a cluster) remain unclassified, here, proposed as A4-clade based on its divergence within A-cluster. The viral haplotypes may link their persistence to geo-climatic conditions and host response. Despite the effectiveness of non-therapeutic interventions in India, multipronged strategies including molecular surveillance based on real-time viral genomic data is of paramount importance for a timely management of the pandemic.

microbiology↗

bioRxiv2018

A novel mutation alters the stability of PapA2 resulting in the complete abrogation of sulfolipids in clinical mycobacterial strains

The analysis of whole genome has revealed specific geographical distribution of Mycobacterium tuberculosis (Mtb) strains across the globe suggestive of unique host dependent adaptive mechanisms in Mtb evolved out of selective pressures. We provide an important correlation of a genome-based mutation to a molecular phenotype across two predominant clinical Mtb lineages of the Indian subcontinent. We have identified a distinct lineage specific mutation-G247C, translating into an alanine to proline conversion in the papA2 gene of Indo oceanic lineage 1 [L1] Mtb strains; restoration of cell wall sulfolipids by simple genetic complementation of papA2 from lineage 3 [L3] or from H37Rv (Lineage 4-L4) attributed the loss of this glycolipid to this specific mutation in Indo oceanic L1 Mtb. Investigation into the structure of Mtb PapA2 revealed a distinct non-ribosomal peptide synthetase (NRPS) C domain conformation with an unconventional presence of Zinc binding motif. Surprisingly, the A83P mutation did not map to either the catalytic centre in the N-terminal subdomain or any of the substrate binding region of the protein. On the contrary, the inherent ability of mutant PapA2 to form insoluble aggregates and molecular simulations with the Wt/ mut PapA2 purports an important role for the surface associated 83rd residue in protein conformation. The present study demonstrates the importance of a critical structural residue in the papA2 protein of Mtb and helps establish a link between observed genomic alteration and its molecular consequence in the successful human pathogen Mtb.

molecular biology↗