Discover, rate, and discuss biology preprints.

Odon is a fast, interactive viewer for spatial proteomics data that outperforms existing tools in loading and rendering large-scale imaging datasets.

The paper introduces a linear-time algorithm to compute ultrabubbles in bidirected pangenome graphs, achieving significant speedups over existing methods.

The paper introduces a method using metabolite fraction libraries to improve NMR metabolomics quantification by reducing spectral overlap and enabling accurate metabolite identification.

This study uses network pharmacology and molecular docking to explore Vernolac's anticancer mechanisms, identifying key phytochemicals and pathways involved in apoptosis, immune modulation, and chemoradiosensitization.

This study examines how undersampling techniques affect hyperparameter selection for non-linear dimensionality reduction methods in chemical space visualization, showing reduced computational costs and improved local structure preservation.

Agentomics is an autonomous system that generates ML models for biomedical tasks, outperforming existing agentic systems and human-engineered models on 11/20 datasets.

This study investigates calcium-dependent conformational changes in TG2 and their role in celiac disease using molecular dynamics and nsSNP analysis.

This paper introduces a Nextflow pipeline to evaluate disease module detection methods, showing variability in results due to method, network, and seed differences.

This paper introduces BaySeq-Y, a Bayesian method to directly quantify mosaic loss of chromosome Y from sequencing data, showing improved accuracy over existing methods.

mirCCC infers microRNA-mediated cell-cell communication from single-cell RNA sequencing by detecting coordinated gene expression changes and models miRNA activity and transfer capacity.

This paper develops a deep learning framework for genomic ancestry prediction that is generalizable across diverse populations and interpretable using attribution techniques.

CPS maps physical coordinates to high-fidelity spatial transcriptomics using a multi-scale context distillation strategy, achieving state-of-the-art imputation and denoising performance.

The paper introduces WxS-QC, a quality control pipeline for human WGS and WES data, and provides an open test dataset to evaluate QC methods.

seq2ribo predicts ribosome location profiles from RNA sequences using a hybrid simulation and machine learning approach, achieving high correlation with experimental data.

MintCNA improves CNA detection in single-cell multi-omics data by integrating deep learning and multivariate change-point detection, outperforming existing methods in simulations and showing strong correlation with bulk sequencing.

Graph-KIR estimates KIR gene copy numbers and calls 7-digit alleles from WGS data with high accuracy, outperforming existing tools.

The paper evaluates host-specific linear B-cell epitope prediction models across human, mouse, and non-human primate datasets, showing significant performance differences and highlighting the need for host-aware approaches.

ProMeta improves disease diagnosis and prediction from plasma proteomics using a meta-learning framework that overcomes data scarcity with biological priors and interpretable features.

MiRformer predicts miRNA-mRNA interactions and localizes binding/cleavage sites using a dual-transformer framework, achieving state-of-the-art performance on benchmark datasets.

CROP predicts CRISPR-Cas9 frameshift rates using a context-aware, feature-independent method, outperforming existing approaches across multiple datasets.