Search bioRxiv⌕ Search

Biology subjects

Kumar, S.

Publications and source records attributed to Kumar, S..

At least 127 records · Page 7Linked to original sources

Common origin for effector and regulatory Follicular and Tissue-Adapted CD4+ T cells in Non-Small Cell Lung Cancer

Tumor-invaded lymph nodes (LNs) serve as critical hubs for anti-tumor immunity, yet their role in orchestrating immune responses remains poorly understood. Using integrated single-cell RNA sequencing, T cell receptor sequencing, and chromatin accessibility profiling, we analyzed CD4+ T cells from matched blood, tumor-invaded LNs, and tumors of treatment-naive non-small cell lung cancer patients. We identified distinct immunological landscapes across these compartments. Compared to blood, tumor-invaded LNs and tumors were enriched for follicular regulatory T cells (Treg-Tfr), conventional T cell subsets with Tfh-like characteristics (Tconv-Tfh and Tconv-CXCL13), and tissue-resident memory Tregs (Treg-Trm). These populations share a BATF-dependent transcriptional program that governs T-cell activation and tissue adaptation, while simultaneously engaging distinct, subset-specific regulatory networks. Integrative TCR-RNA analysis revealed that tumor-reactive, neoantigen-specific T cell clones were enriched within these subsets and demonstrated extensive LN-tumor clonal sharing, indicating active recirculation between compartments. Through clonal coupling analysis and trajectory inference, we uncovered that Treg-Tfr cells function as multipotent progenitors that bifurcate into tissue-resident Treg-Trm or into ex-Tregs adopting a Tfh-like CXCL13+ ewector phenotype. Remarkably, follicular CD4+ T subsets from LNs and tumor were transcriptionally and epigenetically similar and localized to analogous germinal center niches. These findings establish tumor-invaded LNs as functional extensions of the tumor microenvironment that generate and maintain tumor-reactive CD4+ lineages. The identification of tissue-resident Treg-Tfr plasticity reveals a critical developmental checkpoint that could be therapeutically targeted to redirect immunosuppressive programs toward anti-tumor ewector responses.

immunology↗

In search of nonlipogenic ABCA1 inducers (NLAI): precision coregulator TR-FRET identifies diverse signatures for LXR ligands

APOE4, the major genetic risk factor for Alzheimers disease (AD), and ABCA1, required for lipidation of APOE are gene products of the liver X receptor (LXR) receptor. LXR agonists have been validated in animal models as therapeutics for AD, atherosclerosis, and many other diseases. Clinical progress has been thwarted by unwanted hepatic lipogenesis. Structurally diverse LXR ligands were profiled in coregulator TR-FRET (CRT) assays analyzing ligand-induced coactivator recruitment, coactivator selectivity, corepressor dissociation, and LXR isoform selectivity. A multiplex CRT assay was developed to measure synchronous ligand-induced displacement of corepressor by coactivator. Potency for coactivator recruitment to LXR{beta} correlated with induction of ABCA1 in human astrocytoma cells. Correlation with lipogenic activation of sterol response element (SRE) in hepatocarcinoma cells, was more complex. CRT response was diverse revealing ligands with theoretical full agonist, partial agonist, antagonist, and inverse agonist, and other signatures within the same chemical series, suggesting the scope for precision CRT to guide nonlipogenic LXR agonist design.

pharmacology and toxicology↗

Integrated Multi-analytical Framework for Comprehensive Characterization of Lignocellulosic Hydrolysates

Lignocellulosic hydrolysates are promising second-generation feedstocks for sustainable biomanufacturing. However, biomass compositional complexities and a lack of analytical framework, particularly for lignin fraction (LF), have limited holistic characterization and impeded downstream valorization. In this study, we present an integrated multi-analytical framework for in-depth compositional profiling of spruce sawdust hydrolysate (SSH). Combining high-performance anion-exchange chromatography and liquid chromatography, sugars and inhibitors were quantified, with monomeric arabinose (78.8{+/-}0.6 g/L) and acetic acid (14.9{+/-}0.1 g/L) as predominant components, respectively. Additionally, we developed a workflow to quantify and evaluate lignin structures and interlinkages, determining factors in biomanufacturing using a unique spectroscopic-chromatographic approach enabling a comprehensive understanding of the LF. UV-Vis spectroscopy quantified a total aromatic content of 58.2{+/-}0.1g/L, while quantitative 31P NMR clarified the hydroxyl group distribution in the LF. 1H-13C HSQC NMR provided structural identification of lignin units and interconnecting linkages, and gel permeation chromatography showed the molecular weight distribution. Gas chromatography-mass spectrometry identified and quantified 38 lignin-derived monomers and extractives, with benzoic acid (2.7{+/-}0.2 mg/L) as the most abundant. This integrated approach provides holistic biomass compositional insights and cross-validation across methods, overcomes the limitations of a single technique, and delivers a robust workflow to evaluate the biomass suitability for bio-valorization. ABSTRACT GRAPHIC O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=109 SRC="FIGDIR/small/675780v1_ufig1.gif" ALT="Figure 1"> View larger version (38K): org.highwire.dtl.DTLVardef@db6eceorg.highwire.dtl.DTLVardef@47b9b3org.highwire.dtl.DTLVardef@f0a85aorg.highwire.dtl.DTLVardef@1bcb66c_HPS_FORMAT_FIGEXP M_FIG C_FIG SYNOPSISThis study develops an integrated multi-analytical workflow to comprehensively profile biomass hydrolysate for downstream valorization.

bioengineering↗

Identification of donor landraces for early vegetative stage salinity tolerance in wheat by employing multivariate selection indices

Soil salinity (EC of [≥]4 dS m-1) is a major abiotic stress leading to decreased crop production due to osmotic, ionic, and cellular ionic imbalance balance. The current study aims to identify salinity-tolerant wheat landraces by investigating morpho-physiological traits. Three hundred fifty wheat landraces and check lines were evaluated for salt stress (15 dS m-1) tolerance in a hydroponic set-up. The clustering analysis categorized 350 genotypes into four control and 15 dS m-1 conditions clusters. Across the multi-trait genotype-ideotype distance index (MGIDI), Smith Hazel Index (SHI), and Stress Susceptibility Index (SSI), seven genotypes were consistently identified as tolerant among the top 150. In comparison, 12 genotypes were consistently classified as least tolerant. These identified tolerant landraces serves as important resources and novel salt tolerance donors for wheat. The tolerant*sensitive, tolerant* tolerant check, crosses can be used for generating RIL populations to delineate genetic basis of salt tolerance of newly identified lines.

physiology↗

Cell migration sculpts evolutionary dynamics favouring therapy resistance in lung cancer

The evolution of cancer undermines the long-term efficacy of therapy. In this study, we combine experimental, computational, and clinical data analysis to investigate factors influencing the competition of subclones in lung cancer. Lineage tracing reveals unexpected variation in the long-term fate of neutral subclones, with subclones arising near the edge of tumours being favoured. Low levels of cell migration and cell mixing lead to high cell densities in the interior of the tumour that suppress proliferation. Using agent-based modelling and in silico analysis we inferred the extent of cell mixing in human tumours from the TRACERx lung cancer study. This reveals correlations between Epithelial to Mesenchymal Transition (EMT), stromal fibroblasts and levels of cell mixing. Experimental analysis confirms that both TGF{beta}-driven EMT and stromal fibroblasts reduce the variability in subclone fate and promote subclone mixing. Moreover, mixing favours clonal sweeps by subclones resistant to therapy-induced cell killing. Together, these analyses demonstrate that EMT and stromal fibroblasts sculpt tumour evolution by promoting cell mixing and thereby favour the rapid dominance of therapy resistant subclones.

cancer biology↗

Inflammasomes and DNA Damage Orchestrate Divergent Pyroptotic Fates in Neutrophils

The understanding of pyroptotic cell death and its mechanistic insights in neutrophils remains less explored and ambiguous. This study analyzes neutrophil pyroptotic responses under different conditions and dissects distinct pyroptosis-associated events, including IL-1{beta} release and cell death. We decipher the signalling pathways underlying these responses and distinguish them from other cell death mechanisms. Interestingly, we observe IL-1{beta} release and enhanced survival in neutrophils in response to LPS-primed ATP-treatment, in contrast to a combination of IL-1{beta} release with cell death in macrophages. While LPS-primed nigericin-treated neutrophils exhibit IL-1{beta} release and cell death, characterized by nuclear rounding, cell swelling/ ballooning, plasma membrane pore formation, and subsequent cell rupture, confirming the occurrence of pyroptotic cell death. While nigericin alone triggers cytokine uncoupled pyroptosis. Intriguingly, these phenomena do not induce under other death programs, including apoptosis, NETosis. This provides opportunity to unravel the regulation of these pyroptotic responses in neutrophils. Data observed reveal the role of NLRP3 and context-dependent caspase -1 & 11 in IL-1{beta} secretion. While, DNA damage and caspase-7 & 9 regulate pyroptotic death. Intriguingly, forced DNA damage by ATM kinase inhibitor mitigated inflammasome activation and IL-1{beta} release, while spur death. Furthermore, this study identifies perinuclear actomyosin forces driving nuclear rounding and pyroptotic death. Both murine neutrophils exposed to bacteria and human neutrophils treated with nigericin exhibit these pyroptotic processes, highlighting their broad relevance. This study depicts decision whether to secrete cytokines or undergo pyroptotic cell death emphasizing the regulatory role of DNA damage-cytoskeletal axis beyond inflammasome activation. Moreover, LPS and bacteria induced acute lung injury in mice displays nuclear rounding presenting pyroptotic neutrophils and enhanced NLRP3, Caspase-11 activation, and IL-1{beta} release. Together, this study defines intriguing crosstalk of NLRP3, caspases, DNA damage, and actomyosin forces that orchestrate divergent inflammatory fates, potentially leading to opportunities for targeted strategies for combating inflammation. HIGHLIGHTSO_LINeutrophils undergo pyroptotic cell death with nigericin and bacteria, but not with ATP. C_LIO_LILPS priming is essential for IL-1{beta} secretion, while dispensable for cell swelling/ballooning and death. C_LIO_LINLRP3 and Caspase 1/11 regulate IL-1{beta} secretion, while DNA damage and Caspase-7/9 control pyroptotic cell death. C_LIO_LIPerinuclear actomyosin signalling regulates nuclear rounding during pyroptotic cell death. C_LI GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=174 HEIGHT=200 SRC="FIGDIR/small/673694v1_ufig1.gif" ALT="Figure 1"> View larger version (43K): org.highwire.dtl.DTLVardef@102da3forg.highwire.dtl.DTLVardef@46838borg.highwire.dtl.DTLVardef@16648faorg.highwire.dtl.DTLVardef@413bc2_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

Genome based analysis of Antibacterial Biosynthetic Clusters in Lactiplantibacillus plantarum C6 and Exploration of their Natural Small Molecules as anti-biofilm in Methicillin-Resistant Staphylococcus aureus

This study presents the complete genome characterization of Lactiplantibacillus plantarum C6, a strain isolated from Indian dairy cheese, using Illumina NovaSeq sequencing. The assembled genome (3.22 Mb, 44.5% GC) comprised 3,076 coding sequences, 59 tRNAs, 10 rRNAs, and 2 CRISPR arrays. Phylogenomic and ANI analyses confirmed its identity within the L. plantarum clade (>99% similarity with NMGL2 and DMDL 9010). Functional annotation revealed genes enriched in carbohydrate metabolism (10.7%), stress response, and host-adaptation pathways, supporting its probiotic potential. Bacteriocin biosynthetic gene clusters were identified, including those encoding PlnE, PlnF, PlnJ, PlnK, and PlnN, indicating the strains ability to produce class II plantaricins. A RiPP cluster encoding a cyclic uberolysin-like peptide was also detected, with structural similarity to known lanthipeptides such as Streptococcin A, Nisin Q, and Lacticin 3147 (Tanimoto scores 0.93-1.0), suggesting antimicrobial relevance. CAZy analysis revealed 102 carbohydrate-active enzymes (GHs, GTs), highlighting metabolic flexibility. To evaluate the antibiofilm potential of L. plantarum-derived metabolites, 15 small molecules from cell-free supernatants (CFS) were selected through literature mining and subjected to molecular docking against the MRSA biofilm-associated enzyme poly-{beta}-1,6-N-acetyl-D-glucosamine synthase (encoded by icaA). 2,4-Di-tert-butylphenol (-7.2 kcal/mol) and Indole-3-lactic acid (-7.1 kcal/mol) showed the strongest binding, followed by Cyclo (L-propyl-L-valine) (-6.8 kcal/mol) and DL-4-Hydroxyphenyllactic acid (-6.4 kcal/mol), indicating promising inhibition of MRSA biofilm synthesis. Organic acids like acetic and lactic acid showed weaker interactions but may contribute synergistically through acidification. Overall, L. plantarum C6 combines robust probiotic features, genomic safety, and antimicrobial potential, supported by bacteriocin gene clusters and effective antibiofilm metabolites, highlighting its application in functional foods and novel antimicrobial development.

genomics↗

Sensory Compression as a Unifying Principle for Action Chunking and Time Coding in the Brain

The dorsolateral striatum (DLS) supports diverse time-sensitive behaviors--action chunking, duration estimation, and motor timing--yet no single framework is able to explain all of these phenomena. Here, we propose that the massive convergence of cortical projections onto the striatum provides such a framework. We develop a corticostriatal neural network model in which a recurrent cortical module communicates with a recurrent striatal module through a low-dimensional, noisy bottleneck, with the whole system trained via reinforcement learning. Across three DLS-associated tasks, compression produces a consistent computational motif whereby cortex provides low-dimensional control signals while the striatum generates stable, time-encoding dynamics. This separation gives rise to chunking behavior with action slipping, intensity-biased duration judgements with stimulus-modulated time coding, and stereotyped motor timing programs. Perturbation of the compressed cortical signal causally shifts behavior while preserving sequential structure in striatum activity. In sum, our results unify information-theoretic and dynamical systems perspectives of basal ganglia function to link anatomical compression to a variety of temporally-sensitive sensorimotor behaviors implicated in the DLS.

neuroscience↗

UALCAN Mobile, an app for cancer proteogenomic data analysis.

Cancer is a complex disease affecting various organs and is a major cause of death worldwide. During cancer initiation, disease progression, and tumor metastasis, various genomic and proteomic alterations are observed. Recent technological advances have led to the generation of large amounts of molecular data, including genomics and transcriptomics. These large-scale datasets can be utilized to analyze and identify sub-class-specific cancer biomarkers and targets. However, there is a need for the development of user-friendly tools for large-scale data analysis, disseminating the analyzed data in a visualizable format to cancer researchers with no programming skills. We developed UALCAN, a comprehensive platform that allows users to integrate disparate data to better understand the genes, proteins, and pathways perturbed in cancer and make discoveries of potential biomarkers and targets. In the current study, we describe the development of the UALCAN Mobile application (app) that will provide cancer transcriptomic data obtained from The Cancer Genome Atlas (TCGA) project to evaluate protein-coding gene expression based on various stratifications, including stage, grade, race, gender, and molecular-subtypes across over 30 types of cancers. In addition, the UALCAN mobile provides data analysis options for epigenetic changes due to DNA promoter methylation and Clinical Proteomic Tumor Analysis Consortium (CPTAC) cancer proteomic data. The app provides access to large cancer molecular datasets on the go. To find changes in the expression of causative genes and proteins and to identify biomarkers and therapeutic targets, UALCAN mobile app will be extremely valuable. The "UALCAN Mobile" app is free to use and can be downloaded from both the iOS/Apple and the Android Play Store and has been downloaded over 100 times in each of iOS and android app stores. Key PointsO_LICancer, a complex disease affecting various organs, is a major cause of death worldwide. C_LIO_LIMany molecular alterations, gene and protein expression changes lead to the initiation of cancer, disease progression, and tumor metastasis. C_LIO_LIThe available large genomic and proteomics datasets can be utilized to analyze and identify sub-class-specific cancer biomarkers and targets with innovative tools. C_LIO_LIUALCAN Mobile application (app) will provide gene expression, promoter DNA methylation, and protein expression across cancers based on various stratifications, including stage, grade, and molecular subtypes across over 30 types of cancers. C_LIO_LIUALCAN Mobile app is free and can be downloaded from both iOS/Apple and the Android Play Store. C_LI

bioinformatics↗

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↗

Repurposing Melatonin in dual-mode for Wilson disease therapy as a Copper Chelator and an antioxidant agent

Loss-of-function mutations in copper-ATPase ATP7B underlie Wilson disease (WD), a disorder characterized by hepatic copper accumulation and severe hepato-neuropathology. Existing chelation therapeutics remove excess copper but lack intrinsic antioxidant capacity and frequently cause systemic toxicity. Here we evaluate melatonin, an FDA-approved indoleamine with antioxidant and putative metal-chelating activity, as a candidate therapeutic for WD. In ATP7B-/- hepatocytes, melatonin restored copper-induced reactive oxygen species (ROS) to basal levels, reduced apoptosis twofold, and attenuated Nrf2 nuclear translocation leading to reduction of Hemoxygenase-1 abundance. Live-cell ratiometric analysis of GSSG/GSH using GRX1-roGFP2 expressed in melatonin-treated ATP7B-/- hepatocytes revealed a significant reduction in intensity-ratio, indicating an effective mitigation of copper-induced glutathione oxidation. Isothermal calorimetric titration revealed moderate Cu2+ affinity (K ITC=4.54 x 103 M-{superscript 1}), rationalized by MD-simulations showing an interaction energy of 18.5 x 10-3 kcal{middle dot}mol-{superscript 1} via amide-Cu{superscript 2} coordination. In-cellulo studies also revealed that copper-induced vesicularized ATP7B reinstates to Golgi in melatonin-treated hepatocytes. In vivo, melatonin treatment reduced copper-induced oxidative stress in zebrafish embryos and lowered copper burden in Caenorhabditis elegans WD model. Our studies revealed that encapsulation of melatonin within an engineered polymeric nanocapsules having dithiol linkers, susceptible to cleavage by GSH, extended melatonins circulatory half-life ten-fold and enhanced its ROS-scavenging efficacy three-fold relative to free melatonin. This work introduces a unique dual-function therapeutic strategy that integrates antioxidant activity with copper chelation, simultaneously addressing copper overload and redox imbalance. Repurposing melatonin, with its established clinical safety, offers rapid and cost-effective translational pathway toward WD-therapy while providing a generalizable platform for redox- and metal-associated disorders.

cell biology↗

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↗

Affinity Maturation and Light-Chain-Mediated Paratope Diversification Anticipate Viral Evolution

A key goal of vaccinology is to train the immune system to combat current pathogens while simultaneously preparing it for future evolved variants. Understanding factors contributing to anticipatory breadth, wherein affinity maturation against an ancestral strain yields neutralization capacity against evolved variants, is therefore of great importance. Here, we investigated the mechanism of anticipatory breadth development in a public antibody family targeting the functionally restricted ACE2 binding site on SARS-CoV-2. IGHV3-53/66 antibodies isolated from memory B cells of infection-naive individuals vaccinated with the ancestral Wuhan-strain mRNA vaccine frequently neutralized evolved Omicron variants and contained several hallmark mutations previously shown to enhance neutralization breadth. Comparative analyses with antibodies from Omicron breakthrough infections revealed that breadth-associated patterns of somatic hypermutation emerged independently of variant exposure. However, Omicron infection had a marked impact on light chain pairing frequencies, suggestive of variant-imposed selection of favorable light chains. Analysis of available IGHV3-53/66 antibody structures complexed with SARS-CoV-2 receptor binding domain (RBD) clarified these findings; convergent somatic mutations on the heavy chain largely refined contacts with invariant RBD residues, while light chain pairings shifted epitopes to avoid steric challenges posed by Omicron mutations. These findings support a model of anticipatory breadth with three key elements: (1) targeting of a functionally restricted epitope, (2) affinity maturation to establish an affinity buffer, and (3) variable chain pairing to generate paratope diversity. These elements each serve to compensate for a distinct consequence of viral mutagenesis, offering a mechanistic framework for anticipating viral evolution. Graphical Summary O_FIG O_LINKSMALLFIG WIDTH=157 HEIGHT=200 SRC="FIGDIR/small/672735v1_ufig1.gif" ALT="Figure 1"> View larger version (51K): org.highwire.dtl.DTLVardef@1e4e893org.highwire.dtl.DTLVardef@1610bddorg.highwire.dtl.DTLVardef@520909org.highwire.dtl.DTLVardef@1cf4edc_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

Transgenic Cowpea Expressing Synthetic BtCry1Ab Confers High Resistance to Legume Pod Borer (Maruca vitrata)

Cowpea, an important food legume crop, suffers significant yield losses due to insect pests, particularly the legume pod borer (Maruca vitrata). Narrow genetic base and lack of sufficient resistance to M. vitrata in existing cowpea germplasm, make them less effective in managing M. vitrata populations. To address this challenge, a transgenic approach is explored. In this study, we develop transgenic cowpea with enhanced resistance to M. vitrata by overexpressing a synthetic cry1Ab gene, driven by constitutive CaMV35S viral promoter, in an Indian cowpea cultivar. High expression of Cry1Ab was detected in leaves and pods. Bioassays with M. vitrata and Helicoverpa armigera larvae demonstrated significant resistance of transgenic lines to these lepidopteran pests. The transgenic lines exhibited reduced pod damage, decreased larval feeding, and increased insect mortality compared to non-transgenic controls. NMR analysis demonstrated absence of any new metabolites in transgenic plants. Furthermore, these transgenic lines showed no penalty on growth and development. The transgenic lines exhibited no observable phenotypic drawbacks, indicating their promise for sustainable agricultural practices. Our findings suggest that these transgenic cowpea lines could offer broad-spectrum protection against these pests, providing an effective pest management solution.

plant biology↗

Mechanistically explainable AI model for predicting synergistic cancer therapy combinations

This study introduces a Large Language Model (LLM)-based framework that combines drug combination data with a knowledge graph to predict synergistic oncology drug combinations with mechanistic insights. Using a retrieval-augmented generation (RAG) approach, over 50,000 in vitro drug pair assay results and 1,631 human clinical trial or preclinical test entries were integrated to enhance predictive accuracy and explainability. Validation achieved an F1 score of 0.80, demonstrating the frameworks potential to streamline drug discovery and improve translational strategies in cancer treatment.

cancer biology↗

Integrative comparative transcriptomics using cultivated and wild rice reveals key regulators of developmental and photosynthetic progression along the rice leaf developmental gradient

A comprehensive understanding of gene regulatory networks and the key regulators underlying developmental and physiological progression along the leaf developmental gradient is crucial for optimizing photosynthetic competence. Comparisons of developmental and photosynthetic features across successive leaf developmental stages revealed pronounced differences between wild rice (Oryza australiensis) and three cultivated rice accessions. Global gene expression profiling identified three major transcriptional phases across leaf stages: a predominance of developmental genes at SAM+Pi (initiating primordia) and P3; genes for photosynthetic transition at P3 and P4; and enriched core photosynthetic genes at P4 and P5. O. australiensis showed a more prominent expression of developmental and photosynthetic genes than cultivated accessions across leaf stages. Multivariate analysis further supported a distinct transcriptional landscape in O. australiensis compared with cultivated accessions. Integration of gene expression with species-specific variations in regulatory sequences, derived from synteny-anchored orthology analysis, generated stage-resolved gene regulatory networks and identified key transcription factors (TFs) mediating the developmental and physiological progression. A cross-species comparison of TF-target regulatory networks revealed extensive stage- and accession-dependent rewiring of the regulatory networks of key TFs, with strong regulatory divergence in O. australiensis. Gene silencing of two candidate TFs, OsDOF8 and OsARID2, altered the rice photosynthetic competence with accession-specific effects. Taken together, the study highlights promoter-driven regulatory divergence as a potential mechanism underlying developmental and photosynthetic differences among the rice accessions. The resource is available through an interactive public database, Rice DEV-LEAF (https://nipgr.ac.in/DEV-LEAF/), enabling exploration of gene expression dynamics and regulatory networks across the rice leaf developmental gradient.

plant biology↗

Federated Learning for ICU Mortality Prediction: Balancing Accuracy and Privacy in a Multi-Hospital Setting

Intensive care units (ICUs) manage critically ill patients whose clinical outcomes rely on timely and accurate decision-making. Predictive modeling using electronic health records (EHRs) has shown promise in forecasting adverse events such as in-hospital mortality. However, constructing robust and generalizable models typically requires large-scale, multi-institutional datasets. Privacy regulations such as HIPAA and GDPR make centralized data aggregation across hospitals challenging, posing a barrier to collaborative healthcare research. In this study, we explore a privacy-preserving approach for ICU mortality prediction using Federated Learning (FL), which enables distributed model training without sharing sensitive patient data. Leveraging a curated subset of the eICU Collaborative Research Database--a multi-center dataset of over 200,000 ICU admissions--we simulate a real-world federated learning scenario using data from three distinct hospitals comprising 5,000 ICU stays. We design a federated logistic regression framework and evaluate its performance against independently trained local models. Our findings indicate that the federated model consistently outperforms local baselines, achieving improved accuracy and AUC while maintaining strict data privacy. Our work delivers a reproducible and scalable framework for privacy-preserving healthcare AI and demonstrates the practical feasibility of deploying federated models for critical outcome prediction in ICU settings.

bioinformatics↗

Chromosome-Scale, Telomere-to-Telomere Assembly of Winged Bean (Psophocarpus tetragonolobus (L.) DC.) Genome Reveals Lipid Biosynthetic Hotspots and Gene Family Dynamics

Winged bean (Psophocarpus tetragonolobus) is a nutritionally rich but cytologically and genomically underexplored legume with potential for climate-resilient agriculture. We report a telomere-to-telomere, chromosome-scale genome assembly ([~]697.69 Mb; N50 = 85.98 Mb; [~]98.25% of the cytometric estimate) of the cultivar AKWB1 using a combination of PacBio HiFi, Illumina, BioNano, and Hi-C technologies. Cytogenetics revealed a diploid karyotype (2n = 18) with exclusively metacentric chromosomes, establishing the first chromosomal ideotype map for the species. The high-quality assembly anchors 98.28% of bases to nine pseudochromosomes and includes 53,745 annotated protein-coding genes, with 97.6% BUSCO completeness. Repeat elements comprise [~]60% of the genome, dominated by LINEs, LTRs, and DNA transposons. KEGG and GO annotations reveal diverse biosynthetic capabilities across metabolic and stress-response pathways. Comparative genomics with nine legumes demonstrates conserved chromosomal synteny with Glycine max, absence of recent whole-genome duplication, and a contraction-dominated gene family history. Lipid metabolism analysis identified 750+ genes across 12 KEGG pathways, including fatty acid biosynthesis, elongation, and triacylglycerol assembly. Chromosomes 2, 3, and 5 showed enrichment of lipid biosynthetic loci, suggesting evolutionary hotspots. This reference genome positions P. tetragonolobus as a valuable model for studying genome evolution and engineering underutilized legumes for oil yield, protein enrichment, and climate adaptability.

genomics↗