Discover, rate, and discuss biology preprints.

The paper introduces a framework to assess whether in silico perturbations using foundation models reflect biological responses or confounding factors.

DigiAra designs Arabidopsis mutants with enhanced microbial resistance using a simulation-scoring-screening pipeline and hybrid gene interaction modeling.

The paper identifies non-random positional overlap between common germline polymorphisms and somatic cancer mutations, suggesting intrinsically vulnerable genomic loci.

StainX provides a GPU-accelerated framework for batch stain normalization in histopathology images, achieving significant speed improvements over CPU and other GPU implementations.

scATrans annotates single-cell differential expression as transcription- or stabilization-weighted using unspliced RNA, without additional experiments, and recovers mechanism-specific biology from standard scRNA-seq data.

The paper compares True-5mC methylation entropy analysis using nanopore sequencing to bisulfite sequencing in kidney cancer and mouse brain, showing bisulfite introduces systematic biases affecting biological interpretation.

RiboRep predicts codon-resolved ribosome density by integrating RNA sequences and ribosome occupancy signals using a cross-modal transformer framework, achieving strong performance across multiple species.

This paper introduces Vir3D, a structure-aware deep learning model that predicts influenza antigenicity and improves vaccine strain selection by incorporating 3D structural information from amino acid sequences.

This paper introduces REFCON, a deep-learning model for reference-free copy number inference in single-cell tumor transcriptomes, showing improved accuracy and robustness compared to existing methods.

PlantCAD2 is a DNA foundation model pre-trained on 65 angiosperm genomes, showing improved performance in evolutionary conservation and chromatin accessible region prediction compared to existing models.

PhysioMap connects physiological traits to causal relationships and predicts responses in metabolic diseases, showing agreement with references after excluding discordant cases.

The paper introduces FIDDL, a tool to distinguish genuine interspecific introgression from artifacts by using depth-matched negative controls and locus-level tests.

The paper introduces SYNTAX, a method to analyze sequence motifs in the androgen receptor interactome, revealing key motifs and their organization across different compartments and time points.

This paper introduces RGAI, a geometric deep-learning framework for cross-domain analysis of spatial transcriptomics, enabling alignment of latent tissue states across diverse datasets.

The paper improves metabolite effect estimation by incorporating external biological knowledge through a Bayesian hierarchical framework, showing consistent gains in both simulated and real datasets.

LagCI identifies temporal causal relationships in dense multi-omics time series by analyzing lag-correlation profiles and filtering spurious associations.

Mango enables interactive correlation analysis of tabular biological data to distinguish biases from causal relationships in genomic and transcriptomic datasets.

The paper identifies cell types regulated by circulating miRNAs linked to asthma exacerbations using scRNA-seq and validates findings in GEO datasets.

This study evaluates the recovery of CRC-associated microbiome biomarkers using in silico spike-ins, revealing limitations in detection and quantification across different profiling workflows.

SIEVE is a deep-learning framework for variant and gene prioritization in exome studies, achieving strong discrimination and recovering known associations across diseases.