A toolkit for modeling <i>cis</i>-regulatory logic of enhancers at large scale [new] Deciphers genomic regulatory code, offering data-driven insights into cell-type specific enhancer function from single-cell chromatin accessibility atlases.

Linking Genetic Risk to Disease-Relevant Cellular States via Metacell-Informed Modeling with ICePop [new] ...by associating disease w/ metacells, finds het. signals in specific cell states in cell types for testable hypothesis.

OncoMORPHIA: An Integrated Web Platform for Interactive 3D Visualization and Functional Annotation of Cancer Mutations [new] ...unifies 3D protein structs, clin. data, drug-Tx, surv. anal., & AI int. to simplify cancer genomics.

Several multiple sequence alignment perturbation methods enhance AlphaFold3 sampling of alternative protein states [new] Methods improve AF3's capture diverse protein conformations proving useful for dynamic biological understanding.

Methods for Molecular Recognition Computing [new] Leverages softmax/Boltzmann equivalence & specific architectural conditions to define a Foundation Model for diverse biological molecular recognition.

Machine Learning-Driven Antigen Selection Reveals Conserved T-Cell Targets for Broad Coronavirus Vaccination [new] ID'ing conserved non-spike T-cell tgts, developed into multiepitope mRNA vaccines eliciting broad CoV immunity.

Anonymized Somatic Tumor Twins (STTs) enable open genome data sharing and use in research and clinical oncology [new] Anonymize T-N DNA for privacy, pres. somatic variants for open genomic data sharing in cancer res. & oncology.

LigandForge: A Web Server for Structure-Guided De Novo Drug Design [new] offers an accessible web server for structure-guided de novo ligand gen, assembling fragments into synthesizable, drug-like compounds without local software.

muat: portable transformer-based method for tumour classification and representation learning from somatic variants [new] provides this method robust software for portable deployment across diverse HPC/SPE envs, using WGS/WES data.

Left atrial flow and thrombosis risk from 4D CT contrast dynamics by physics-informed neural network and indicator dilution theory [new]

Rectifying AI-generated protein structure ensembles for equilibrium using physics-based computations [new] ...by seeding phys-based sims from AI data and reweighting trajectories to a consistnt equilibrium ensemble for a force field.

Evaluation of direct strain field prediction in bone with data-driven image mechanics (D2IM-Strain) [new] ...predicts bone strain directly from XCT images, removing noisy numerical differentiation and keeping computational efficency.

A Generative Neuro-Symbolic AI for Protein Sequence Design [new] ...synergizes deep learning & automated reasoning, encoding fitness landscapes into probabilistic models explored to design sequences with long-range dependencies.

Towards a Cytometry Foundation Model: Interpretable Sample-level Predictive Modelling via Pretrained Transformers [new] via PTD Trans., from varied markers for interpretable S-L pred. & IDing key cell subsets for biol. validation.

StrataBionn: a neural network supervised classification method for microbial communities [new] ...accurately classifies microbial community states using an ANN capturing complex, non-linear taxa profiles, applicable to body sites.

TF-IDF k-mer-based Classical and Hybrid Machine Learning Models for SARS-CoV-2 Variant Classification under Imbalanced Genomic Data [new]

Correcting Preprocessing Bias in Sparse Chromatin Contact Data Enables Physically Interpretable Reconstruction of Genome Architecture [new]

Decoding antibiotic modes of action from multimodal cellular responses [new] ...via integrated proteomic, structural, and growth profiles from *E. coli* to classify MoA and identify novel drug candidates.

Interpretable Electrophysiological Features of Resting-State EEG Capture Cortical Network Dynamics in Parkinsons Disease [new] ...by using distinct standard and dynamical features that discern PD disease and medication states, revealing network alterations.

A Novel Multi-view Mixture Model Framework for Longitudinal Clustering with Application to ANCA-Associated Vasculitis [new] ...integrates static/longitudinal data via a mixture model with ODE-based temp patterns for interpretable disease subgrp discovery.

annbatch unlocks terabyte-scale training of biological data in anndata [new] ...by enabling out-of-core training directly from disk-backed datasets, bypassing data access limits and preserving scverse ecosystem compatibility.

Evaluating Deep Surrogate Models for Knee Joint Contact Mechanics Under Input-Limited Conditions [new] ...revealed a hybrid model's robustness, but optimal model choice under min. inputs depended on tasks incl. error, risk region, or hotspot localization.

QuantumXCT: Learning Interaction-Induced State Transformation in Cell-Cell Communication via Quantum Entanglement and Generative Modeling [new] Core: QCs learn unitary transforms, mapping non-int. to int. cell states for de novo disc.

Histone Modification Metapeaks are Epigenetic Landmarks Predictive of Cell State [new] Identified in a vast ChIP-seq atlas consistently enriched genomic regions. Landmarks predict cell's state and highlight key regulatory regions.

When Multimodal Fusion Fails: Contrastive Alignment as a Necessary Stabilizer for TCR--Peptide Binding Prediction [new] Ensures intra-entity modality consistency preventing noisy structural data from destabilizing multimodal fusion.

A structure-informed deep learning framework for modeling TCR-peptide-HLA interactions [new] integrates physchem, seq, struct features to predict TCR-p-HLA int., aiding cancer immunotherapy & autoimm. research by identifying key Ags.

SHOT-CCR: Biologically guided adversarial training for test-time adaptation in cellular morphology [new] decouples morph signals & exp confounds via cell-invariant gradient reversal, improving generaliz. on diverse Cell Painting data

DESPOT: Direction-Enhanced Scoring POTentials [new] ...is an anisotropic KBP framework modeling ligand atom type probabilities at discrete spatial positions around protein atoms, captures directional preferences and steric exclusion.

Benchmarking Agentic Bioinformatics Systems for Complex Protein-Set Retrieval: A Coccolithophore Calcification Case Study [new] Eval. 3 agents on retrieval/cat. of coccolith calc proteins assessing rel., cat., repeat.

The U-method: Leveraging expression probability for robust biological marker detection [new] Identifies unique genes by comparing expression probability within cluster vs highest in others, prioritizing detection cons. for cell pops.

Generating and navigating single cell dynamics via a geodesic bridge between nonlinear transcriptional and linear latent manifolds [new]

Digital Twins for Fungal Computing: Viable XOR Regimes, Parameter Inference, and Waveform-Guided Rediscovery [new] exploring XOR regimes, inferring biophys. params, and refining models via waveform match aid computer-aided design.

Mapping the mammalian dark metabolome by in vivo isotope tracing [new] ...identifies biosynthetic origins for thousands of unidentified metabolites, uncovering new metabolic families and age-related isoprenoid changes.

AI-Driven Reconstruction of the Research Paradigm for Phase Separation in Membraneless Organelle [new] transforms phase separation research from binary prediction to mechanistic understanding and novel MLO discovery.

DeepTrio: Variant Calling in Families Using Deep Learning [updated] It learns inheritance patterns, de novo rates, and sequencing/mapping error weighing directly from joint child-mother-father sequence data.

Protein and genomic language models uncover the unexplored diversity of bacterial immunity [new] Identify & characterize novel bacterial immune systems, revealing diverse defense strategies & genetic architectures.

DefensePredictor: A machine learning model to discover prokaryotic immune systems [new] ...for comprehensive identification of novel host defense genes and pathways in diverse prokaryotes.

CardamomOT: a mechanistic optimal transport-based framework for gene regulatory network inference, trajectory reconstruction and generative modeling [new]

mRNA-GPT: A Generative Model for Full-Length mRNA Design and Optimization [new] ...jointly optimizes all mRNA regions using RL to design sequences for desired properties like half-life and translation efficiency.

CREsted: modeling genomic and synthetic cell-type-specific enhancers across tissues and species [new] ...an efficient toolbox to analyze, model, and design these specific regulatory elements.

Generative machine learning unlocks the first proteome-wide image of human cells [new] by simulating thousands of protein images from sparse stains to create virtual cells for spatial organization modeling.

Object Detection Techniques for Live Monitoring of Amoeba in Phase-Contrast Microscopic Images [new] ...by developing DL models for auto-det., improving bio-imaging accuracy, reducing manual effort, & enabling low-int. obs.

De novo Folding Mechanisms of Lasso Peptides [new] Uncovered are an uphill landscape, low folding probability, and high entropy cost, with loop stability and confinement guiding distinct rotaxane pathways.

Baktfold: Sensitive protein functional annotation across the microbial tree of life using structural information [new] Annotates microbial proteins via struct info, converts seqs to 3Di tokens for DB searches to ID hypothetical ones.

emb2dis: a novel protein disorder prediction tool based on ResNets, dilated convolutions & protein language models [new] Predicts disorder from seq using pLM embeds & novel arch. w/ ResNets/dilated convs, expanding context capture.

De novo design of a peptide ligand for specific affinity purification of human complement C1q [new] via AlphaFold2-designed cyclic peptides immobilized on streptavidin for robust C1q capture, even from plasma.

SSPSPredictor: A Sequence and Structure based Deep Learning Model for Predicting Phase-Separating Proteins [new] ESM-2 seq & GVP struct data predict PSPs, revealing IDPs have higher LLPS tendency & linking pathog. variants to ↑ LLPS.

GenoBERT: A Language Model for Accurate Genotype Imputation [new] It's a transformer-based, reference-free framework tokenizing phased genotypes, using self-attention to capture LD dependencies, overcoming ancestry bias.

Genetic algorithms for multi-omic feature selection: a comparative study in cancer survival analysis [new] introduces an iterative multi-view multi-objective algorithm identifying compact Bx panels evaluating within-layer signals & cross-layer interactns.

Latent-Y: A Lab-Validated Autonomous Agent for De Novo Drug Design [new] It autonomously executes full antibody design, from literature review to lab-ready sequences, extensible to other biologics like peptides.