Discover, rate, and discuss biology papers.

Random forests on sequence and 2D chemistry pick the favored protein for a molecule, or the favored molecule for a protein, at 0.71 to 0.78 accuracy on unseen compounds.

This study evaluates CTGAN and TGAN for generating synthetic longitudinal CVD data, finding CTGAN offers better fidelity than TGAN but has lower performance in temporal fidelity and for younger age groups.

The paper validates and analyzes 12,000 AI-designed CAR-T binders from a global competition, identifying successful designs and factors improving their performance.

msengine improves HDX-MS signal extraction by adaptively adjusting mass and retention-time boundaries, increasing accepted targets and reducing processing time.

This paper introduces RosettaLigand PathFinder, a tool for simulating ligand trajectories through proteins, showing it can reproduce known transport features and mutation effects.

PGRanker improves Mendelian gene prioritization by integrating phenotype data with gene networks using graph learning, achieving higher recall and better ranking of causal genes.

The paper identifies CCT2 as a marker of a translation-activated state in sporadic ALS motor neurons and links it to functional decline.

This paper introduces a Pareto optimization method for masked superstrings to improve the compression of pan-genome k-mer sets, achieving 12-19% better compressibility.

The paper shows deep learning models struggle to generalize cellular perturbation responses across contexts, highlighting the importance of causal mechanisms over distributional similarity.

The paper quantifies how increasing train-test sample relatedness inflates classification performance in biomedical image benchmarks using a microscopy dataset.

SpatialOmicsLab provides an integrated environment with AI tools for spatial transcriptomics analysis, outperforming expert workflows in multiple tasks.

drugMR uses multi-omics data to identify potential drug targets and adverse effects by integrating GWAS and pQTL statistics with various statistical methods.

RASTRO improves unitig graph path selection by using sequence-based models to identify read-supported branching points in HIV-1 and SARS-CoV-2 assemblies.

The paper shows that 8-oxoguanine can create premature-stop transcripts in PAH genes, mimicking pathogenic variants without DNA mutation.

This paper introduces a supervised learning denoising method to identify low-quality labels in TCR specificity data, improving model reliability by filtering noisy examples.

The paper evaluates EviSearch, a clinical evidence-curation system, showing high accuracy and improved results with targeted review.

This study identifies nonlinear gene-expression trajectories in the aging male mouse colon, revealing midlife transitions and a transcriptomic clock based on three genes.

alamem and alasight are tools for fast and memory-efficient local alignment and horizontal gene transfer detection in bacterial genomes.

The paper proposes a weighted estimation method to correct for imperfect mutation detection in tumor sequencing due to intra-tumor heterogeneity, reducing bias and type I error.

FlyAnatoView projects Drosophila scRNA-seq data onto anatomical diagrams to restore spatial context and enable exploration of gene expression patterns across tissues.