Discover, rate, and discuss biology preprints.

GraphTME predicts immune checkpoint inhibitor response by analyzing spatial cell-cell interactions in tumor microenvironments using spatial transcriptomics data.

This study compares RAM-NJ and SAM-NJ phylogenies using protein foundation models to reconstruct species relationships without conventional substitution models or multiple sequence alignments.

The paper introduces a low-temperature zwitterionic HILIC method for LC-MS metabolomics, showing improved peak shapes, resolution, and retention time stability compared to standard methods.

The paper identifies distinct transcriptomic signatures in chronic lung allograft dysfunction phenotypes using bulk RNA sequencing and non-negative matrix factorization.

This paper uses a DNA foundation model to prioritize promoter regulatory variants in colorectal cancer, identifying high-impact variants enriched in cancer-related pathways and GWAS loci.

The paper introduces BIFO, an ontology that specifies biologically admissible directed transformations in biomedical knowledge graphs to enable meaningful signal propagation.

gTranslate predicts prokaryotic translation tables with high accuracy using machine learning, outperforming existing methods and identifying new genetic code variations.

CardioSeg provides an integrated platform for analyzing spatial transcriptomics and nuclear morphology in mouse heart tissue, achieving high accuracy in segmentation and revealing disease-specific changes.

TIPP-SD improves species detection accuracy in microbiomes, especially for low-abundance species and with sequencing errors, while maintaining reasonable speed.

Unimeth improves DNA methylation detection from nanopore reads across species and modification types by using a transformer framework with multi-phase training.

DupyliCate is a Python tool for analyzing gene duplications across species, demonstrating its use on plant and microbial datasets.

BINSEQ introduces binary formats for sequencing data to enable faster parallel processing, reducing analysis time for genome and transcriptome studies.

PFES quantifies the protein context of missense variants, enabling mechanistically transparent interpretation and reclassification of variants of uncertain significance.

C4TUNE predicts photosynthesis kinetic parameters from response curves, enabling efficient simulation and identification of limiting factors in maize genotypes.

This study identifies accelerated aging patterns in 3D genome organization and transcriptome in schizophrenia using Hi-C data and transcriptomic analysis.

CARIBOU is a multi-agent AI system that autonomously performs bioinformatics analyses using HPC environments, adapting and refining workflows across datasets.

Deep-Palm is a deep learning framework that accurately predicts S-palmitoylation sites by integrating sequence, physicochemical, embedding, and structural features, outperforming existing methods.

The paper introduces a computational framework to identify antimicrobial targets by analyzing protein dynamics, showing that few protein-level mechanisms account for most antimicrobial effects.

PhyloStage infers multi-stage pathway-level cancer progression models from tumor phylogenies, aligning with known AML progression and stratifying NSCLC patients by tumor size.

UcTCRp profiles MAIT- and iNKT-associated immune states using TCRβ sequences without requiring single-cell data or antigen-specific markers.