Discover, rate, and discuss biology preprints.

BAT constructs and annotates gene trees to explore functional relationships within gene families using local genomic resources.

The paper introduces MetaFR, a method that links scATAC-seq open-chromatin variation to scRNA-seq gene expression, showing improved performance over existing methods.

MicrobeMS is a MATLAB toolbox for microbial identification using MALDI-ToF MS, enabling database compilation, spectrum analysis, and identification tasks.

The paper validates AlphaFold2-generated protein structures by aligning them with structurally similar proteins, showing success for challenging cases like ABHD16A.

AutoTuneX improves transcript assembly accuracy by predicting optimal parameters for RNA-seq samples.

The paper evaluates deep learning models for predicting protein melting temperatures using proteomics and biophysics datasets, finding that fine-tuned embeddings improve performance on diverse validation sets.

The paper benchmarks the Narrative Velocity framework for next clinical event prediction in MIMIC-IV, showing improved accuracy over non-neural baselines and analyzing the interaction between structured features and embeddings under self-distillation.

LysinFusion improves phage lysin prediction using a hybrid CNN-Transformer model, achieving higher accuracy and fewer false positives than existing methods.

Lense optimizes single-cell omics preprocessing by using language models to select the best pipeline based on visual comparisons of low-dimensional representations.

The paper introduces LAMNr flows for multimodal data analysis, enabling shared latent subspaces and cross-view imputation, with applications in biological imaging.

This paper introduces scPathOT, a pathway-informed domain adaptation method for single-cell RNA-seq data, demonstrating its ability to accurately annotate cell types and uncover functional mechanisms across diverse biological contexts.

pyTrance identifies co-localizing RNAs in spatial transcriptomics data with higher accuracy than existing methods, revealing biologically relevant RNA co-localization patterns in mouse brain tissue.

This paper introduces MT-GEM and STEP to predict discrete structural transformations in metabolites from MS/MS spectra, achieving significant improvements over existing methods.

ProteinFlux predicts protein conformational dynamics influenced by post-translational modifications using a generative framework, outperforming traditional methods in speed and accuracy.

This study evaluates seven differential abundance methods across six datasets, finding that some produce biased p-values under null conditions, while classical tests perform reliably.

InterScale models both local and global cellular interactions in spatial transcriptomics data, revealing spatially restricted differentiation programs and disease-associated tissue remodeling.

This paper introduces a diffusion model for generating realistic human microbiome data, achieving high sparsity preservation and competitive ecological distance metrics.

BioMADE predicts Torsades de Pointes risk using biologically informed molecular representations, outperforming chemical-only models.

This paper uses open-weight large language models to predict synthetic lethal interactions, outperforming random chance and non-LLM methods in reproducing CRISPR screen results.

The paper benchmarks six long-read simulators against Oxford Nanopore WGS data, finding PBSIM3 best reproduces read-level properties but none capture all error features.