Discover, rate, and discuss biology preprints.

ARACoFusion predicts protein-protein interactions in Arabidopsis thaliana using a deep learning model that improves accuracy and reliability through uncertainty-aware techniques and benchmarking against existing predictors.

GAP-MS improves gene prediction accuracy in plant genomes by integrating mass spectrometry evidence, identifying hundreds of previously unannotated protein-coding loci.

The paper benchmarks machine learning classifiers on TEP RNA-seq data to classify cancer types, finding Multinomial Logistic Regression achieves the highest performance.

The paper investigates if structural distance to the tRNA-binding interface improves missense variant pathogenicity prediction, finding that existing models like AlphaMissense and EVE already capture this information.

The paper introduces FlatStat, a tool to estimate sample tilt from 3D fluorescence stacks, and applies it to public data to assess tilt variability across projects.

This paper introduces a vision-based model for predicting CNV pathogenicity using track-based spatial representations, achieving improved performance on benchmarks and clinical datasets.

GAE-Δ uses a graph autoencoder to analyze multi-omics data and predict clinical outcomes while identifying gene network rewiring associated with cancer progression.

This paper develops AMLS, an evolutionary learning method combined with SVR, to identify a miRNA signature for improved survival prediction in AML, achieving high prognostic accuracy.

ARGus co-assembles metagenomes to generate MAGs, detect ARGs, and analyze virulence factors in environmental samples.

MultiTME integrates spatial and single-cell data using a cycle-consistent deep generative model, revealing a spatially coherent tumor axis in colorectal cancer.

The paper introduces triAMPh, a framework for predicting pathogen-specific antimicrobial peptide activity using biological large language models, showing it matches or outperforms existing methods.

This study applies computer vision tools to maize genomic data to predict genotype-phenotype relationships and identify gene regions associated with traits.

The paper identifies ASXL1 mutations in clonal hematopoiesis as a risk factor for lung cancer, particularly in smokers, using large population-based cohorts.

GDDN connects gene-sets to diseases and therapeutics to prioritize drugs, enabling functional analysis and personalized medicine.

OTTER improves T-ALL molecular characterization by integrating transcriptomic and genomic data using optimal transport, outperforming existing methods and identifying subtype-specific features.

This paper identifies a bias in allosteric site prediction benchmarks and introduces AlloDyn, a dynamic-aware framework that improves prediction performance by incorporating conformational dynamics.

The paper constructs an IID-specific knowledge graph from biomedical literature to support disease ontology refinement and drug repurposing.

maxiM/Ze processes LC-HRMS metabolomics data using image recognition to automate feature detection, annotation, and alignment with minimal user input.

Tandem detects, classifies, and quantifies bacterial tandem gene duplications in reference genomes and population samples with high accuracy.

OryzaG3 is a rice-specific DNA language model pretrained on a pangenome, achieving competitive performance on a genomic benchmark while improving inference efficiency.