Discover, rate, and verify biology preprints.

This study evaluates if language models improve combinatorial molecular search for ionizable lipids compared to classical methods under budget constraints, finding no significant improvement.

ALBIS generates synthetic spatial transcriptomics data across multiple resolutions and dimensions to enable benchmarking of computational methods.

The paper predicts cyanobacterial diazotrophy using protein-family content and prioritizes candidate protein families for experimental validation.

tRNAZAP is a deep learning toolkit for analyzing nanopore tRNA sequencing data, achieving high accuracy in segmentation and classification tasks.

This study shows that surrogate variable analysis (SVA) can lead to overfitting and reduced cross-cohort reproducibility in neuropsychiatric gene expression analyses.

broadSeq benchmarks differential gene expression methods and shows DELocal identifies distinct genes by considering genomic neighbourhood, particularly for highly expressed genes.

scGSCA analyzes single-cell RNA sequencing data to detect differential gene-set co-expression across phenotypes, identifying pathway changes missed by bulk methods.

The paper introduces an adaptive minibatch approach for approximate Bayesian inference, improving speed-precision tradeoffs for large, heterogeneous biological datasets.

phyloDS identifies differentially used orthologous splice junctions across species by improving orthology definition, statistical testing, and sequence-level validation.

STARgrid is an open-source pipeline that processes Visium HD sequencing data to generate count matrices, showing high agreement with Space Ranger while improving efficiency and resolution.

AnnZarro enables interactive exploration of cell-by-cell and gene-by-gene relationships in single-cell data through a browser-based viewer linked to AnnData objects.

The paper assesses covariance stationarity in spatial transcriptomics and identifies spatially varying genes using non-stationary covariance models.

UPLiFT improves parameter efficiency in fine-tuning protein language models by optimizing activation space instead of network weights, achieving competitive performance with fewer trainable parameters.

The paper improves a tool for sorting DNA reads in xenograft models, enhancing efficiency for epigenetic analysis.

CLARA uses a language model to make strategy decisions in compound library active learning, outperforming random sampling across multiple protein targets.

The paper evaluates whether molecular models can causally represent chemical variables by comparing intervention effects with random controls and pretraining advantages.

rhythmCI provides an R package for detecting circadian rhythms using harmonic regression confidence intervals, improving rhythmicity analysis in small clinical datasets.

The paper extends Ramachandran diagrams by modeling amino acid and pair torsion-angle distributions conditioned on secondary structure elements, enabling more accurate conformational predictions.

The paper investigates how specific stalk-derived peptides can reactivate signaling in disease-associated PC1 mutants and uses computational simulations to support these findings.

The paper harmonizes Indian gut microbiome cohorts to create a reference atlas and the Indian Gut Health Index, revealing geographic and cohort-specific microbial patterns.