Biology subjects

Gautam, V.

Publications and source records attributed to Gautam, V..

13 recordsLinked to original sources

bioRxiv2026

Scalable Computational Phenomics of Nuclear Morphology

Interpretable computational pathology is constrained by a trade-off between scalable learned representations and biologically explicit phenotypes. Here we present NucXplore, a nuclear phenomics framework that converts each hematoxylin-and-eosin (H&E)-stained nucleus into 129 explicitly defined features spanning morphology, chromatin, intensity, texture, color, and spatial context. A Rust-based implementation accelerates extraction 17.6-fold over the Python reference implementation. Across three histopathology cohorts, NucXplore outperformed classical descriptors and larger deep-learning embeddings while retaining feature attribution. Applied to human skin, hierarchical multiple-instance models resolved eight cellular compartments and decoded age and sun-exposure states, revealing a largely shared aging program with age and cell-specific exposure-dependent recalibration. Conditional flow matching and neural ODE integration transformed cross-sectional phenotypes into model-implied trajectories, uncovering a midlife velocity minimum, late-life re-acceleration, and cell-specific changes in trajectory magnitude and direction. Source-free domain adaptation enabled preliminary transfer of age-associated predictions to an independent hospital cohort. NucXplore thus unifies interpretable representation, high-performance computing, and dynamical modeling for cellular phenotyping from routine histology.

pathology↗

bioRxiv2025

Chemical Dice Integrator (CDI): A Scalable Framework for Multimodal Molecular Representation Learning

The machine learning landscape for molecular property prediction is fragmented, with numerous Featurizers each capturing a narrow, specialized view of chemical structure. This heterogeneity forces a suboptimal choice of representation a priori, limiting model generalizability. We introduce the Chemical Dice Integrator (CDI), a hierarchical framework that unifies six orthogonal molecular representations, physicochemical (Mordred), topological (GROVER), visual (ImageMol), biological (Signaturizer), quantum-mechanical (MOPAC), and linguistic (ChemBERTa), into a single, coherent embedding. The framework consists of CDI-Basic, a two-tiered autoencoder that fuses these modalities, and CDI-Generalised, a Mamba State-Space Model (SSM) that learns a direct, efficient map from SMILES strings to the unified embedding space. Extensive benchmarking across 23 classification (171 tasks) and 10 regression datasets demonstrates that CDI embeddings consistently achieve superior predictive performance compared to individual Featurizers and standard feature aggregation methods. The CDI-Generalised model achieves this performance with exceptional computational efficiency, outperforming deep learning Featurizers in terms of speed and resource overhead. Furthermore, we demonstrate that the CDI embedding is chemically intuitive, allowing for the sensitive distinction of nuanced structural variants, such as chiral enantiomers and kekulized SMILES forms. By bridging multimodal chemical intelligence with scalable, sequence-based inference, CDI offers a strong foundation for molecular machine learning.

bioinformatics↗

bioRxiv2025

Non-Invasive Diagnostic Evaluation of Urinary Exosomal Let-7c Cluster Expression in Bladder Cancer Using Machine Learning Approaches

BackgroundBladder cancer (BCa) diagnosis typically relies on invasive cystoscopy, which is effective but costly and uncomfortable. Urinary microRNAs (miRNAs), especially exosomal ones, are promising non-invasive biomarkers due to their stability in biological fluids and disease specificity. However, challenges such as population variability, methodological inconsistencies and normalization issues hinder their clinical translation, emphasizing the need for innovative approaches to enhance diagnostic performance. ObjectiveTo evaluate the diagnostic potential of urinary exosomal let-7c cluster (let-7c-5p, miR-99a-5p and miR-125b-5p) in BCa patients by integrating miRNA expression data with Machine Learning (ML) models. MethodsUrine samples were collected from 66 participants, including 50 BCa patients and 16 healthy controls (HC). Exosomal miRNAs were isolated and quantified using Quantitative Real-Time-Polymerase-Chain-Reaction (qRT-PCR). Statistical analysis and hypothesis tests were conducted to explore the nature and diagnostic relevance of individual biomarkers. A logistic regression classifier was applied to evaluate both the combined and differential diagnostic capabilities of the selected biomarkers. Accuracy, precision, recall and AU-ROC scores were used to assess model performance. Bioinformatics analysis was performed to identify pathways associated with the features prioritized by the ML models, ensuring their relevance to BCa. ResultsThe result revealed significant differentiation between BCa patients and HC, with miR-99a-5p (p=0.013, AU-ROC=0.71) and miR-125b-5p (p=0.047, AU-ROC=0.64) demonstrating reliable diagnostic performance (let-7c-5p showed weaker discrimination, AU-ROC=0.65, p>0.1). The logistic regression ML model achieved an accuracy of 80.0% (AU-ROC=0.86, recall=100%) in distinguishing cancer from HC and 53.3% (AU-ROC=0.63) when applied to miRNA-only grade classification. When clinical variables were integrated with miRNA expression, performance improved to 73.3% accuracy (AU-ROC=0.61) for high-versus low-grade differentiation. Across Ta-T2, miR-99a-5p displayed relatively better separation, whereas let-7c-5p and miR-125b-5p showed weak stage-related differences. The integration of bioinformatics analysis confirmed the biological relevance of these miRNAs in BCa-related pathways, including PI3K-Akt, p53, NF-{kappa}B and RAS/MAPK signaling, with hub genes such as TP53, MYC, EGFR, and CCND1 identified, further validating the diagnostic utility of the selected biomarkers. ConclusionUrinary let-7c cluster miRNAs demonstrate promising diagnostic potential when analyzed with ML models, offering a non-invasive alternative to conventional methods. These findings highlight the promise of ML-based approaches alongside molecular markers for advancing clinical diagnostics in BCa.

molecular biology↗

bioRxiv2025

SynGlue: AI-Driven Designer for Clinically Actionable Multi-Target Therapeutics

The rational design of protein degraders, such as proteolysis-targeting chimeras (PROTACs), requires the simultaneous optimization of multiple molecular properties, a complex challenge that limits efficient discovery. Here, we introduce SynGlue, a generative artificial intelligence (AI) framework that addresses this challenge through two core modules: data-driven, leveraging large-scale protein-ligand intelligence, and structure-guided, for physics-aware molecular design. SynGlue harness MagnetDB, a curated database of 6.37 million experimental protein-ligand interactions, and couples it with deep learning models that quantitatively predict degradation potency (DC50), maximal degradation (Dmax), and guide ternary-complex-compatible linker design. Benchmarked against 6,935 compounds, SynGlue demonstrates superior performance in relevant pharmacology prediction. To validate SynGlue, we engineered degraders for BRD4 and GSPT1. Our data-driven design for BRD4 yielded compounds with novel warhead scaffolds (<50% warhead similarity with known PROTACs), which proved to be potent degraders in vitro (DC50 = 0.19 nM) and efficacious in vivo in mouse models. Independently, our structure-guided de novo design for GSPT1 produced ultrapotent degraders (DC50 {approx} 0.0011 M) that are also effective both in vitro and in vivo, uncovering a new oncogenic dependency. By unifying data-driven and physics-aware design, SynGlue establishes a generalizable AI framework for the rapid development of clinically relevant protein degraders, with principled extension to other multi-target modalities.

bioinformatics↗

bioRxiv2025

MutAIverse: An AI-Powered, Mechanism-backed Platform for Discovering Novel DNA Adducts and their precursor Genotoxins

Genotoxin exposure leads to DNA adduct formation, potentially causing mutations if unrepaired. Current DNA adductomics platforms or analytical workflows are limited by incomplete spectral libraries, reliance on experimentally validated adducts, limited cellular contexts, and inefficient computational methodologies. We introduce MutAIverse, an advanced AI-driven DNA adductomics analysis platform that overcomes these limitations by leveraging intracellular mechanistic modeling of genotoxin bioactivation to construct a comprehensive DNA adduct library and offers advanced spectral mapping. MutAIverse integrates experimentally validated and chemically valid putative DNA adducts, enabling interpretable retrograde tracking to parental genotoxins. Validation of MutAIverse against experimental MS/MS datasets demonstrated the detection of both known and novel adducts. Furthermore, application to in-house generated DNA adductomics data from tissue biopsies of smokeless tobacco-induced head and neck cancer patients revealed selective enrichment of both novel and known DNA adducts; a subset of them was validated using MS/MS analysis. Collectively, MutAIverse provides a robust, end-to-end, and interpretable platform for advanced DNA adductomics analysis.

bioinformatics↗

bioRxiv2024

Language model-guided anticipation and discovery of unknown metabolites

Despite decades of study, large parts of the mammalian metabolome remain unexplored. Mass spectrometry-based metabolomics routinely detects thousands of small molecule-associated peaks within human tissues and biofluids, but typically only a small fraction of these can be identified, and structure elucidation of novel metabolites remains a low-throughput endeavor. Biochemical large language models have transformed the interpretation of DNA, RNA, and protein sequences, but have not yet had a comparable impact on understanding small molecule metabolism. Here, we present an approach that leverages chemical language models to discover previously uncharacterized metabolites. We introduce DeepMet, a chemical language model that learns the latent biosynthetic logic embedded within the structures of known metabolites and exploits this understanding to anticipate the existence of as-of-yet undiscovered metabolites. Prospective chemical synthesis of metabolites predicted to exist by DeepMet directs their targeted discovery. Integrating DeepMet with tandem mass spectrometry (MS/MS) data enables automated metabolite discovery within complex tissues. We harness DeepMet to discover several dozen structurally diverse mammalian metabolites. Our work demonstrates the potential for language models to accelerate the mapping of the metabolome.

bioinformatics↗

bioRxiv2024

DNA methylation and demethylation driven regulation of sepsis

During dysregulated inflammation and sepsis, there is a sudden surge of cytokines, such as TNF alpha, IL6, and IL10, which disrupts the body homeostasis. This sudden and rapid increase in cytokine gene expression cannot be explained by the conventional central dogma mechanism. DNA methylation is one of the most significant epigenetic modifications. DNA methylation is mostly associated with the suppression of gene activity, and DNA demethylation is associated with the activation of gene activity. There are numerous transcription factors, such as CREB1, c-FOS, AP-1, IRF1, and EGR1, that play pivotal roles in regulating inflammation. Thus, it was hypothesized that changes in the DNA methylation patterns of cytokine promoter regions or in the promoter regions of these specific transcription factors might be the potential reasons for the increase in cytokine levels during sepsis. During endotoxin stimulation, DNA methyltransferases were suppressed, and DNA methylation levels were altered at the global level. Bisulfite sequencing revealed no change in the DNA methylation patterns of the TNF-alpha, IL6 and IL10 cytokine promoters. The CREB1 and c-FOS transcription factor promoters were demethylated after LPS stimulation and in clinical sepsis samples. Overall, this study highlights the importance of the role of DNA methylation in sepsis.

molecular biology↗

bioRxiv2024

Introducing identification probability for automated and transferable assessment of metabolite identification confidence in metabolomics and related studies

Methods for assessing compound identification confidence in metabolomics and related studies have been debated and actively researched for the past two decades. The earliest effort in 2007 focused primarily on mass spectrometry and nuclear magnetic resonance spectroscopy and resulted in four recommended levels of metabolite identification confidence - the Metabolite Standards Initiative (MSI) Levels. In 2014, the original MSI Levels were expanded to five levels (including two sublevels) to facilitate communication of compound identification confidence in high resolution mass spectrometry studies. Further refinement in identification levels have occurred, for example to accommodate use of ion mobility spectrometry in metabolomics workflows, and alternate approaches to communicate compound identification confidence also have been developed based on identification points schema. However, neither qualitative levels of identification confidence nor quantitative scoring systems address the degree of ambiguity in compound identifications in context of the chemical space being considered, are easily automated, or are transferable between analytical platforms. In this perspective, we propose that the metabolomics and related communities consider identification probability as an approach for automated and transferable assessment of compound identification and ambiguity in metabolomics and related studies. Identification probability is defined simply as 1/N, where N is the number of compounds in a reference library or chemical space that match to an experimentally measured molecule within user-defined measurement precision(s), for example mass measurement or retention time accuracy, etc. We demonstrate the utility of identification probability in an in silico analysis of multi-property reference libraries constructed from the Human Metabolome Database and computational property predictions, provide guidance to the community in transparent implementation of the concept, and invite the community to further evaluate this concept in parallel with their current preferred methods for assessing metabolite identification confidence.

biochemistry↗

bioRxiv2023

Artificial Intelligence uncovers Evolutionarily Conserved Intracellular Allosteric Modulators of GPCR-Gα interface.

Endogenous intracellular allosteric modulators of GPCRs remain largely unexplored, with limited binding and phenotype data available. This gap arises from the lack of robust computational methods for unbiased cavity identification, cavity-specific ligand design, synthesis, and validation across GPCR topology. Here, we developed Gcoupler, an AI-driven generalized computational toolkit that leverages an integrative approach combining de novo ligand design, statistical methods, Graph Neural Networks, and Bioactivity-based ligand prioritization for rationally predicting high-affinity ligands. Using Gcoupler, we interrogated intracellular metabolites that target and regulate the GPCR-G interface (Ste2p-Gpa1p), affecting pheromone-induced programmed cell death in yeast. Our computational analysis, complemented by experimental validations, including genetic screening, multi-omics, site-directed mutagenesis, biochemical assays, and physiological readouts, identified endogenous hydrophobic metabolites, notably sterols, as direct intracellular allosteric modulators of Ste2p. Molecular simulations coupled with biochemical signaling assessment in site-directed Ste2p mutants further confirmed that metabolites binding to GPCR-G obstruct downstream signaling, possibly via a cohesive effect. Finally, by utilizing isoproterenol-induced, GPCR-mediated human and neonatal rat cardiac hypertrophy models, we observed that elevated metabolite levels attenuate hypertrophic response, reinforcing the evolutionary relevance of this mechanism.

bioinformatics↗

bioRxiv2022

Visualization of Incrementally Learned Projection Trajectories for Longitudinal Data

Longitudinal studies that continuously generate data enable the capture of temporal variations in experimentally observed parameters, facilitating the interpretation of results in a time-aware manner. We propose IL-VIS (Incrementally Learned Visualizer), a new machine learning pipeline that incrementally learns and visualizes a progression trajectory representing the longitudinal changes in longitudinal studies. At each sampling time point in an experiment, IL-VIS generates a snapshot of the longitudinal process on the data observed thus far, a new feature that is beyond the reach of classical static models. We first verify the utility and correctness of IL-VIS using simulated data, for which the true progression trajectories are known. We find that it accurately captures and visualizes the trends and (dis)similarities between high-dimensional progression trajectories. We then apply IL-VIS to longitudinal Multi-Electrode Array data from brain cortical organoids when exposed to different levels of Quinolinic Acid, a metabolite contributing to many neuroinflammatory diseases including Alzheimers disease, and its blocking antibody. We uncover valuable insights into the organoids electrophysiological maturation and response patterns over time under these conditions.

bioinformatics↗

bioRxiv2021

Artificial Intelligence uncovers carcinogenic human metabolites

The genome of a eukaryotic cell is often vulnerable to both intrinsic and extrinsic threats due to its constant exposure to a myriad of heterogeneous compounds. Despite the availability of innate DNA damage response pathways, some genomic lesions trigger cells for malignant transformation. Accurate prediction of carcinogens is an ever-challenging task due to the limited information about bona fide (non)carcinogens. We developed Metabokiller, an ensemble classifier that accurately recognizes carcinogens by quantitatively assessing their electrophilicity as well as their potential to induce proliferation, oxidative stress, genomic instability, alterations in the epigenome, and anti-apoptotic response. Concomitant with the carcinogenicity prediction, Metabokiller is fully interpretable since it reveals the contribution of the aforementioned biochemical properties in imparting carcinogenicity. Metabokiller outperforms existing best-practice methods for carcinogenicity prediction. We used Metabokiller to unravel cells endogenous metabolic threats by screening a large pool of human metabolites and predicted a subset of these metabolites that could potentially trigger malignancy in normal cells. To cross-validate Metabokiller predictions, we performed a range of functional assays using Saccharomyces cerevisiae and human cells with two Metabokiller-flagged human metabolites namely 4-Nitrocatechol and 3,4-Dihydroxyphenylacetic acid and observed high synergy between Metabokiller predictions and experimental validations.

bioinformatics↗

bioRxiv2021

Pathogenic bacteria grown in iron and zinc milieu exhibit magnetotaxis

The growing number of multiple drug resistant (MDR) bacteria and the dwindling pipeline of new antibiotics are driving us towards a post-antibiotic era in which even common infections would become difficult to treat. To address this, an antibiotic-free strategy that can combat multiple bacteria is recommended. Most of the proposed approaches nevertheless have several limitations, including bacterial targeting. To overcome such limitations, the proposed strategy employs the bacterial machinery to self-destruct. Herein, the biosynthesis of magnetic nanoparticle (MNP) is reported for the first time in multiple pathogenic bacteria, including MDR bacteria. The intracellular MNPs composed of superparamagnetic zinc ferrites were formed in presence of iron and zinc precursors. Exposure of the treated bacteria/biofilms to an alternating magnetic field (AMF) exhibited hyperthermia (5-6{degrees}C) and a dramatic decrease in bacterial viability, suggesting the MNPs therapeutic potential. Likewise, the bacteria existing in vivo biosynthesize the MNPs by mining these elements from the host. To determine its therapeutic efficacy, the infected tissues were exposed directly to AMF. A 3-4 log reduction in bacterial burden, as compared to antibiotics treatment, confirmed the significance of using naturally existing MNPs to combat bacterial infections. The proposed broad-spectrum approach can therefore aid in overcoming the challenges facing anti-bacterial therapies.

microbiology↗

bioRxiv2019

Extreme genome selection towards complete antimicrobial resistance in a nosocomial strain of Stenotrophomonas maltophilia complex

We report first complete genome sequence and analysis of an extreme drug resistance (XDR) nosocomial Stenotrophomonas maltophilia that is resistant to the mainstream drugs i.e. trimethoprim/sulfamethoxazole (TMP/SXT) and levofloxacin. Taxonogenomic analysis revealed it to be a novel genomospecies of the Stenotrophomonas maltophilia complex (Smc). Comprehensive genomic investigation revealed fourteen dynamic regions (DRs) exclusive to SM866, consisting of diverse antibiotic resistance genes, efflux pumps, heavy metal resistance, various transcriptional regulators etc. Further, resistome analysis of Smc clearly depicted SM866 to be an enriched strain, having diversified resistome consisting of sul1 and sul2 genes. Interestingly, SM866 does not have any plasmid but it harbors two diverse super-integrons of chromosomal origin. Apart from genes for sulfonamide resistance (sul1 and sul2), both of these integrons harbor an array of antibiotic resistance genes linked to ISCR (IS91-like elements common regions) elements. These integrons also harbor genes encoding resistance to commonly used disinfectants like quaternary ammonium compounds and heavy metals like mercury. Hence, isolation of a novel strain belonging to a novel sequence type (ST) and genomospecies with diverse array of resistance from a tertiary care unit of India indicates extent and nature of selection pressure driving XDRs in hospital settings. There is an urgent need to employ complete genome based investigation using emerging technologies for tracking emergence of XDR at the global level and designing strategies of sanitization and antibiotic regime.\n\nImpact StatementThe hospital settings in India have one of the highest usage of antimicrobials and heavy patient load. Our finding of a novel clinical isolate of S. maltophilia complex with two super-integrons harbouring array of antibiotic resistance genes along with antimicrobials resistance genes indicates the extent and the nature of selection pressures in action. Further, the presence of ISCR type of transposable elements on both integrons not only indicates its propensity to transfer resistome but also their chromosomal origin suggests possibilities for further genomic/phenotypic complexities. Such complex cassettes and strain are potential threat to global health care. Hence, there is an urgent need to employ cost-effective long read technologies to keep vigilance on novel and extreme antimicrobial resistance pathogens in populous countries. There is also need for surveillance for usage of antimicrobials for hygiene and linked/rapid co-evolution of extreme drug resistance in nosocomial pathogens. Our finding of the chromosomal encoding XDR will shed a light on the need of hour to understand the evolution of an opportunistic nosocomial pathogen belonging to S. maltophilia.\n\nRepositoriesComplete genome sequence of Stenotrophomonas maltophilia SM866: CP031058

genomics↗