Discover, rate, and discuss biology preprints.

This paper introduces TreeOwen, algorithms for computing Owen values in tree-based ensembles to provide structured, multi-resolution explanations for grouped features in scientific applications.

GEOMeta automates metadata curation for GEO transcriptomic data, generating standardized annotations for 600,000 human bulk RNA-seq samples and benchmarking LLMs for annotation quality.

MSMICA improves metabolite identification in untargeted metabolomics by integrating mass spectrometry data, retention time, and biological evidence, achieving high accuracy across mammalian datasets.

This paper evaluates circulating microbial DNA as a cancer biomarker, finding that observed patterns are largely artifacts of sequencing and analysis methods rather than biological signals.

PlantOmicsGWAS provides a reproducible framework for plant GWAS and genomic prediction, supporting both linear and pan-genome references to improve analysis of structural variations.

The paper uses sparse autoencoders to analyze BiRNA-BERT, revealing biologically meaningful features and improving RNA-type classification.

PerturbTrace evaluates how AI co-scientist agents use feedback in perturbation discovery, finding that high recall does not imply effective feedback use.

The paper identifies and validates defense-related proteins in Solanaceae species using phylogenomics, structural modeling, and machine learning, revealing lineage-specific adaptations in immune responses.

PASTRI infers stage-specific cell-state transition rates from annotated cell lineages, revealing rate-limiting steps and attractor states in development and disease.

This paper introduces HAPBDB, a haplotype-based breeding framework that translates regional haplotypes into breeding decisions, demonstrating its effectiveness in lettuce for QTL pyramiding.

This paper introduces CIFM, a generative virtual tissue model that simulates transcriptional dynamics under therapeutic conditions to design immunotherapy strategies for cancer and autoimmune diseases.

OTRec is a deep learning model that recommends druggable disease-target associations, outperforming existing methods in both retrospective and cross-validation settings.

bulk2scDiff generates single-cell RNA-seq data from bulk profiles using a diffusion model, showing good reconstruction of cell populations in cancer datasets.

This study benchmarks 12 AI-based molecular generation models for drug design across 176 protein-ligand systems, finding no single method optimizes all criteria and proposing a hybrid approach.

The paper derives a closed-form formula for determining the minimum sketch size in MinHash Jaccard estimation, accounting for both noise and precision constraints.

The paper identifies old-like genes in certain aquatic eukaryotes, suggesting a potential antiviral defense system evolutionarily related to prokaryotic anti-phage systems.

This paper evaluates eight in silico perturbation methods across four datasets, finding that only two consistently detect TF-to-pathway regulation, highlighting method choice's impact on results.

MetaPilot improves metaproteomics analysis by using genome-aware search-space refinement, enhancing peptide identification in DDA and DIA datasets.

This paper evaluates 3D Zernike descriptors for antibody structural clustering and repurposing, finding they identify more functional pairs than SPACE2 while maintaining precision.