Barathi Lenin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Barathi Lenin has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Aarti Ravindran
Department of Biotechnology, Bhupat and Jyoti Mehta School of Biosciences, Indian Institute of Technology Madras
2 shared publications - Inderjeet Tyagi
Centre for DNA Taxonomy, Molecular Systematics Division, Zoological Survey of India, Kolkata, India
2 shared publications - Karthik Raman
Department of Biotechnology, Bhupat and Jyoti Mehta School of Biosciences, IIT Madras
2 shared publications - Shradha Sharma
Bhupat and Jyoti Mehta School of Biosciences, Indian Institute of Technology (IIT) Madras
2 shared publications - Abhor Gupta
Robert Bosch Centre for Data Science and AI, Wadhwani School of Data Science and AI (WSAI), Indian Institute of Technology (IIT) Madras, Chennai, India
1 shared publication - Balaraman Ravindran
IIT Madras, India
1 shared publication
Latest Publications
- Open Access
PURE: Policy-guided Unbiased REpresentations for structure-constrained molecular generation
Structure-constrained molecular generation (SCMG) generates novel molecules that are structurally similar to a given molecule and have optimized properties. Deep learning solutions for SCMG are limited in that they are pre-disposed towards existing knowledge, and they suffer from a natural impedance mismatch problem due to the discrete nature of molecules, while deep learning methods for SCMG often operate in continuous space. Moreover, many task-specific evaluation metrics used during training often bias the model towards a particular metric -”metric-leakage”. To overcome these shortcomings, we propose Policy-guided Unbiased REpresentations (PURE) for SCMG that learn within a framework simulating molecular transformations for drug synthesis. PURE combines self-supervised learning with a policy-based reinforcement-learning (RL) framework, thereby avoiding the need for external molecular metrics while learning high-quality representations that incorporate an inherent notion of similarity specific to the given task. Along with a semi-supervised training design, PURE utilizes template-based molecular simulations to better explore and navigate the discrete molecular search space. Despite the lack of metric biases, PURE achieves competitive or superior performance than state-of-the-art methods on multiple benchmarks. Our study emphasizes the importance of reevaluating current approaches for SCMG and developing strategies that naturally align with the problem. Finally, we illustrate how our methodology can be applied to combat drug resistance, by identifying sorafenib-like compounds as a case study.
Open AccessPURE: Policy-guided Unbiased REpresentations for structure-constrained molecular generation
Structure-constrained molecular generation (SCMG) generates novel molecules that are structurally similar to a given molecule and have optimized properties. Deep learning solutions for SCMG are limited in that they are pre-disposed towards existing knowledge, and they suffer from a natural impedance mismatch problem due to the discrete nature of molecules, while deep learning methods for SCMG often operate in continuous space. Moreover, many task-specific evaluation metrics used during training often bias the model towards a particular metric -”metric-leakage”. To overcome these shortcomings, we propose Policy-guided Unbiased REpresentations (PURE) for SCMG that learn within a framework simulating molecular transformations for drug synthesis. PURE combines self-supervised learning with a policy-based reinforcement-learning (RL) framework, thereby avoiding the need for external molecular metrics while learning high-quality representations that incorporate an inherent notion of similarity specific to the given task. Along with a semi-supervised training design, PURE utilizes template-based molecular simulations to better explore and navigate the discrete molecular search space. Despite the lack of metric biases, PURE achieves competitive or superior performance than state-of-the-art methods on multiple benchmarks. Our study emphasizes the importance of reevaluating current approaches for SCMG and developing strategies that naturally align with the problem. Finally, we illustrate how our methodology can be applied to combat drug resistance, by identifying sorafenib-like compounds as a case study.
- Open Access
Metagenomic Profiling of Drinking Water Microbiomes: Insights into Microbial Diversity and Antimicrobial Resistance
Monitoring microbial components in drinking water is as essential as tracking its chemical composition. Although traditional culture-based methods can offer valuable insight into the morphology, behaviour, and phenotype of an organism, their scope is limited to culturable species. With the advent of high-throughput sequencing, we can now detect a wider range of microbes in any ecosystem, gaining more efficient insights into their functional potential and metabolic capabilities. In this study, we performed metagenomic analyses to fully understand the microbiome of drinking water supplied through public distribution systems in an Indian city. Our findings identified bacteria from the phyla Pseudomonadota, Planctomycetota, Bacteroidota, and Actinomycetota, consistent with previous studies of drinking water microbiomes of other countries. We observed that the temperature of the water samples, even as little as a 5°C increase, influenced the composition and diversity of the microbial communities. No significant correlation was detected between the abundance of microbial species and the metal concentration in the sample. In addition, we traced the distribution of antibiotic resistance genes (ARGs), finding widespread resistance to aminoglycosides, tetracyclines, and macrolides across samples. However, the risk associated with these ARGs was beyond the scope of this study. In summary, this research provides a preliminary analysis of the drinking water microbiome, providing a foundation for future management strategies.
Open AccessMetagenomic Profiling of Drinking Water Microbiomes: Insights into Microbial Diversity and Antimicrobial Resistance
Monitoring microbial components in drinking water is as essential as tracking its chemical composition. Although traditional culture-based methods can offer valuable insight into the morphology, behaviour, and phenotype of an organism, their scope is limited to culturable species. With the advent of high-throughput sequencing, we can now detect a wider range of microbes in any ecosystem, gaining more efficient insights into their functional potential and metabolic capabilities. In this study, we performed metagenomic analyses to fully understand the microbiome of drinking water supplied through public distribution systems in an Indian city. Our findings identified bacteria from the phyla Pseudomonadota, Planctomycetota, Bacteroidota, and Actinomycetota, consistent with previous studies of drinking water microbiomes of other countries. We observed that the temperature of the water samples, even as little as a 5°C increase, influenced the composition and diversity of the microbial communities. No significant correlation was detected between the abundance of microbial species and the metal concentration in the sample. In addition, we traced the distribution of antibiotic resistance genes (ARGs), finding widespread resistance to aminoglycosides, tetracyclines, and macrolides across samples. However, the risk associated with these ARGs was beyond the scope of this study. In summary, this research provides a preliminary analysis of the drinking water microbiome, providing a foundation for future management strategies.