Discover, rate, and discuss biology preprints.

This work evaluates factors affecting generalization of GNN-based TCR-pMHC binding predictors to unseen peptides and proposes architectural improvements.

The paper develops a stochastic model for affinity maturation and identifies an optimal antigen-concentration protocol that maximizes the probability of reaching a target antibody affinity.

ProtPen combines sequence and structure-based methods to annotate proteins of unknown function across proteomes with >90% accuracy.

BioMaster automates complex bioinformatics workflows using a multi-agent system, improving accuracy, efficiency, and scalability across tasks like RNA-seq and Hi-C processing.

SpacerScope provides a high-speed, high-sensitivity off-target profiling tool for RNA-guided nucleases, accurately recovering all validated off-target sites and detecting complex indel-containing sites.

CoMR improves mitochondrial proteome reconstruction by integrating multiple evidence sources, achieving better performance than standalone predictors in both model and non-model eukaryotes.

The paper develops a pathogenicity prediction model for hemoglobin variants using structure-informed features and ML, showing strong performance compared to AlphaMissense and providing a complete mutational landscape for HbA.

This paper introduces a supervised contrastive learning framework for multi-omics prediction models, showing improved accuracy over single-omics approaches in type 2 diabetes and high-fat diet studies.

The paper introduces MetaSBT, a tool for classifying microbial genomes and MAGs, and presents databases of over 190,000 viral genomes, including many previously unknown species clusters.

plsMD improves plasmid reconstruction from short-read assemblies, outperforming existing tools in recall, precision, and F1 scores on benchmark and novel datasets.

This study investigates how lithocholic acid modulates GPBAR1 activation and G protein engagement through molecular dynamics simulations and structural analysis.

This paper develops a framework combining machine learning and metabolic modeling to predict synergistic cancer therapies influenced by microbiome composition in colorectal cancer.

This study evaluates how large language models perform in-class learning for antibody classification using biological sequences, finding that few-shot prompting with diverse demonstrations achieves competitive results.

This paper shows that focusing on complementarity-determining regions (CDRs) in antibody sequences improves binding affinity prediction using region-aware pretraining strategies.

LncPNdeep classifies long non-coding RNAs using a deep learning model that combines nucleotide and peptide embeddings, achieving high accuracy in human transcript data.

This study benchmarks cell-type deconvolution methods across cross-platform transcriptomic data, finding that log-normal regression methods like SpatialDecon perform most reliably.

The paper quantifies numerical variability in MRI analyses of Parkinson's disease and shows it can significantly affect statistical conclusions and p-values.

This paper introduces a framework for comparing embedding spaces in molecular biology, demonstrating its utility in pre-filtering machine learning tasks and identifying differential patterns across embeddings.

fourSynergy combines multiple 4C-seq tools using gradient-free optimization to improve interaction calling performance across datasets.