Discover, rate, and discuss biology preprints.

RNA-SpecFlow synthesizes histology images conditioned on RNA data, showing that increased detection counts can coincide with reduced segmentation agreement.

The paper uses single-cell RNA-seq and explainable AI to identify transcriptional features distinguishing invasive from non-invasive TNBC cells in a cell line, finding EMT-related and inflammatory genes as key predictors.

The paper presents an end-to-end pipeline for segmenting nanoparticles and cellular structures in 3D electron microscopy data of tumor spheroids, revealing preferential NP clustering near nuclei.

CATFISHR integrates transcriptomic and copy-number data to improve malignant cell identification in scRNA-seq, achieving high accuracy across multiple datasets and identifying previously missed malignant populations.

VenusGenome improves microbial genome representation learning by using a genome-aware masked autoencoder, achieving high performance in phenotype prediction and gene-level tasks.

The paper develops GPCRmd-NMR, an integrative platform combining MD simulations and NMR spectroscopy to analyze GPCR dynamics at unprecedented scale.

The paper identifies brain-specific functions of NDD-associated genes by analyzing gene expression and regulatory mechanisms across tissues.

The paper introduces CSDE, a method that improves differential expression analysis in spatial transcriptomics by incorporating expert feedback to mitigate preprocessing biases.

ArthroVerse compiles and annotates protein families from arthropod-associated microbiomes to explore functional diversity and host-microbe interactions.

CellCompass, an agentic AI system, models cell dynamics from single-cell measurements and improves predictions by integrating new biological constraints and experimental data.

This paper evaluates how unified perturbational modeling improves AI virtual cell predictors by integrating heterogeneous Perturb-seq datasets and achieving better performance across benchmarks.

AltereGO fuses protein language models and structure representations to predict molecular functions, achieving better performance than individual models and some existing methods.

This paper evaluates how different evaluation designs affect cross-donor generalization in single-cell classification, finding that random cell-level splitting overestimates performance compared to donor-held-out evaluation.

The paper introduces a conversational framework using an LLM agent to interrogate proteomics data sets, identifying issues in four LC-MS/MS datasets and demonstrating its feasibility in reanalyzing data.

The paper introduces PIPPA, a pipeline for processing large-scale ATAC-seq data, showing that peak calling performance varies by dataset and recommending parameters for high-confidence chromatin accessibility profiles.

BLOOM generates developmental cell-expression distributions across time using conditional flow matching, outperforming baselines in gene and cell-expression correlations.

MPFIL predicts mutation-induced changes in protein-protein binding affinity by integrating structural and sequential features across multiple levels.

This paper introduces Mimic, a unified neural network architecture for genomic sequence modeling, showing it outperforms existing models in protein-DNA interactions and enhancer activity.

KmerSignificance Score (KSS) prioritizes viral k-mers by integrating strain-distinguishing capacity, mutation impact, and functional importance, achieving superior classification performance and cross-dataset comparability.

clrcycle is a dimensionality reduction method for cyclic data that captures phase and magnitude in biological systems, showing effectiveness in circadian and cell-cycle analysis.