Biology subjects

Badrinarayanan, S.

Publications and source records attributed to Badrinarayanan, S..

4 recordsLinked to original sources

bioRxiv2025

Large Language Model Agent for Modular Task Execution in Drug Discovery

We present a modular framework powered by large language models (LLMs) that automates and streamlines key tasks across the early-stage computational drug discovery pipeline. By combining LLM reasoning with domain-specific tools, the framework performs biomedical data retrieval, domain-specific question answering, molecular generation, property prediction, property-aware molecular refinement, and 3D protein-ligand structure generation. In a case study targeting BCL-2 in lymphocytic leukemia, the agent autonomously retrieved relevant biomolecular information--including FASTA sequences, SMILES representations, and literature--and answered mechanistic questions with improved contextual accuracy over standard LLMs. It then generated chemically diverse seed molecules and predicted 67 ADMET-related properties, which guided iterative molecular refinement. Across two refinement rounds, the number of molecules with QED > 0.6 increased from 34 to 55, and those passing at least four out of five empirical drug-likeness rules rose from 29 to 52, within a pool of 194 molecules. The framework also employed Boltz-2 to generate 3D protein-ligand complexes and provide rapid binding affinity estimates for candidate compounds. These results demonstrate that the approach effectively supports molecular screening, prioritization, and structure evaluation. Its modular design enables flexible integration of evolving tools and models, providing a scalable foundation for AI-assisted therapeutic discovery.

pharmacology and toxicology↗

bioRxiv2025

Interpreting biochemical text with language models:a machine learning framework for reaction extraction and cheminformatic validation

Recent advancements in large language models (LLMs) offer new opportunities for automating the manual curation of biochemical reaction databases from scientific literature. In this study, we present an integrated pipeline that enhances LLM-based extraction of enzymatic reactions with machine learning and cheminformatics-informed validation. Using BRENDA-linked PubMed articles, we evaluate GPT-4s ability to extract reactions and infer missing chemical entities in textual descriptions of enzymatic reactions. Extracted reactions are converted to SMILES and InChI notations before being encoded into molecular fingerprint similarity scores and atom mapping metrics. These cheminformatics metrics are then used to train machine learning classifiers that validate GPT extractions. We employ a Positive-Unlabeled learning approach with synthetic invalid reactions to train various classifiers and assess model performances. The best classifier is then benchmarked on GPT extractions. Our findings show that GPT can accurately infer incomplete reactions and cheminformatics tools can serve as effective predictors of reaction validity. This work demonstrates a scalable framework for automated and reliable curation of enzymatic reaction databases, highlighting the potential of combining LLMs with cheminformatics and machine learning for reliable scientific knowledge extraction. Author SummaryCurating databases of biochemical reactions is a time-consuming and manual task, yet it plays a vital role in advancing research in biology and chemistry. Many scientific articles describe important enzymatic reactions, but often do so in incomplete ways--such as mentioning only the starting molecule or the enzyme, and leaving out the rest. In this work, we explore how recent advancements in artificial intelligence, specifically large language models like GPT, can help extract such information automatically from scientific literature. We show that these models can not only find reactions in text, but also infer missing parts of reactions based on the surrounding context. To make sure these inferred reactions are chemically plausible, we use computational chemistry tools that analyze the structure of the molecules involved. We then train a machine learning model to help us automatically detect which reactions are likely to be valid. This combination of tools offers a new way to speed up and improve how biochemical knowledge is extracted from the growing body of scientific literature. Our study suggests that this kind of automation could help scientists keep biological databases up to date and reduce the burden of manual data entry.

bioinformatics↗

bioRxiv2024

Medial Entorhinal VIP-expressing interneurons receive direct input from Anterior Dorsal Thalamus and are critical for spatial memory

Head-direction (HD) cells are found across several regions in the brain, including the anterodorsal thalamic nucleus (ADN), the subicular complex, and the medial entorhinal cortex (MEC). A fundamental role of head direction cells is to provide input to MEC grid cells, which are thought to translate information about head direction into a metric code for spatial location. However, classic anatomical studies indicate that most thalamic HD projections pass indirectly to the MEC via the post- and para-subiculum, with only a small subset of ADN fibers terminating in the MEC. To further investigate the smaller and direct projection to the MEC, we use rabies-mediated retrograde tracing in mice to determine if this projection explicitly targets a subset of MEC neurons. Our findings reveal that ADN neurons specifically project onto MEC interneurons, with a preference for MEC VIP-expressing cells. Additionally, MEC VIP cells receive input from the hippocampus, the subicular complex, and the retrosplenial cortex - key centers for spatial memory - suggesting a specialized role for MEC VIP cells in spatial memory. Indeed, we find that MEC VIP cells exhibit increased c-Fos expression in a spatial memory task and show that chemogenetic inhibition of these neurons impairs task performance. Together, these data uncover a specific projection of head direction information onto MEC interneurons and confirm that MEC VIP-expressing cells are critical for spatial memory.

neuroscience↗

bioRxiv2020

A characterization of the electrophysiological, morphological and input domains of vasoactive intestinal peptide (VIP) interneurons in the medial entorhinal cortex (MEC)

Circuit interactions within the medial entorhinal cortex (MEC) translate movement into a coherent code for spatial location. Entorhinal principal cells are subject to strong lateral inhibition, suggesting that a disinhibitory mechanism may drive their activation. Cortical Vasoactive Intestinal Peptide (VIP) expressing inhibitory neurons predominantly contact interneurons, providing a local disinhibitory mechanism. Here, we investigate the electrophysiological and morphological properties of VIP cells using in vitro whole-cell patch clamp recordings and use rabies-mediated circuit tracing to discover long-range inputs that may modulate this population in mice. We report physiological and morphological properties of VIP cells that differ across lamina and along the dorsal-ventral MEC axis. Furthermore, we reveal long-range inputs to VIP neurons from regions known to encode proprioceptive and auditory information, including the mesencephalic trigeminal nucleus and superior para-olivary nuclei, respectively. These results characterize the properties of VIP cells and reveal sensory modalities that could drive disinhibition in the MEC.

neuroscience↗