Discover, rate, and discuss biology preprints.

This paper introduces temporal entropy as a measure of dynamic instability to improve psychiatric classification across cohorts using blood transcriptomic data, finding that excluding individuals with unstable transcriptomic dynamics enhances classification performance.

The paper introduces TopRepo, a comprehensive TD-MS spectral repository, and shows it improves proteoform identification and supports DL model training for spectral prediction.

This paper introduces TxConformal, a framework that controls false discoveries in AI-driven therapeutic discovery by adjusting for distribution shifts and providing confidence measures for AI scores.

DCAFA analyzes histological images to identify compositional shifts and feature associations in biomedical data using a regression-based framework.

The paper introduces evo3D, an R package for structure-informed evolutionary analysis, enabling spatial haplotype extraction and diverse downstream analyses in 3D structural contexts.

The paper develops a framework to analyze performance plateaus in H&E-to-molecular biomarker translation, distinguishing methodological limitations from intrinsic information ceilings.

The paper establishes the relationship between Gaussian process priors and regularized regression in sequence-to-function mapping, showing how to construct regularizers that correspond to explicit Gaussian process priors and characterize implicit function space priors.

AURA is a scalable machine learning framework for predicting antibiotic resistance in Mycobacterium tuberculosis, identifying novel resistance-associated loci.

The paper uses ESM-2 embeddings to represent bacterial genomes, recovering taxonomy and enabling geometry-aware triage in foodborne pathogen surveillance.

The paper identifies distinct intrinsic sequence patterns between HIV-1 DNA and RNA across subtypes using a k-mer-based approach and presents a new tool for genomic analysis.

This paper introduces a Bayesian method to align transcript dynamics across different time scales and species, enabling comparative analysis of gene expression patterns.

The paper introduces three physics-inspired computational methods for spatial transcriptomics to analyze oral mucosal carcinogenesis, identifying a dysplasia-restricted MRB cluster with autocrine signaling and a density-asymmetric niche.

The paper introduces metagenomic-deepFRI, a framework that integrates protein structure models into functional annotation pipelines, improving GO term prediction confidence and coverage in metagenomic datasets.

AnewSampling is a generative model that accurately samples all-atom equilibrium distributions of biomolecular interactions, outperforming prior methods and enabling efficient exploration of conformational landscapes.

GRNPred is a graph transformer model pretrained on gene expression data to infer gene regulatory networks, achieving high AUROC and AUPRC scores on benchmark datasets.

This study uses machine learning on RNA-seq data to identify sex-specific gene panels for bladder cancer biomarkers, achieving high accuracy in distinguishing cancer from non-tumor samples.

The paper identifies a bimodal distribution of peptide collision cross sections, showing two stable conformations, and proposes a two-valued predictor to improve peptide identification in proteomics.

This study designs DNA tweezers that selectively bind S100A4 and midkine, biomarkers of glioblastoma invasiveness, using computational methods and molecular docking.

The paper introduces GenPept-Curated-2025, a benchmark dataset for antimicrobial peptide prediction with controlled homology and class balance.

The paper presents an open-source workflow for multi-target compound prioritization in drug discovery, integrating docking, ADME descriptors, and chemoinformatics for flaviviruses.