Discover, rate, and discuss biology papers.

Translating cells between mouse, macaque and human scRNA-seq with a CycleGAN, without ortholog mapping, preserved developmental time structure, imputed missing states and produced more realistic synthetic cells than scGAN.

DECIPHER improves cell-type deconvolution across molecular modalities by combining disentangled representation learning with prototype-based methods, showing robust performance in simulated and real datasets.

CHACAM improves cell identity annotation in C. elegans scRNA-seq data by integrating cell-cell interactions and gene expression to resolve ambiguous lineages.

The paper identifies sex-specific differences in ADAR editing patterns during SARS-CoV-2 infection, showing distinct expression and editing levels between males and females across infection stages.

MycorrhizaTracer automates Sanger DNA sequence processing and classification for fungi and plants, improving efficiency and accuracy in biodiversity studies.

The paper introduces Murmurent, an agentic AI environment that supports biomedical research collaboration by providing infrastructure, specialized agents, memory, traceability, SOP enforcement, and multi-user capacity.

This study benchmarks preprocessing workflows for spatial transcriptomics data across multiple platforms and identifies platform-specific recommendations to improve downstream analysis.

DisenTE improves translation-efficiency matrix completion by modeling sparse sequence-context patterns, achieving higher accuracy than existing methods.

The paper uses AI to compare biological information from UniProt records and their cited literature for human kinases, finding literature provides better coverage for certain topics.

This study evaluates how well three structural alphabets retain fold-level information compared to full-structure comparisons using TM-scores as a reference.

The paper introduces sum-h2, a fast framework for evaluating genetic discovery from deep learning-derived phenotypes using heritability measures, and establishes a Genetic Discovery Arena for benchmarking.

The paper introduces a latent variable mediation framework to analyze the gut microbiome's indirect effects on clinical outcomes through compositional data structures.

The paper develops classifiers to detect modified 16S rRNA sequences that pass SILVA database quality controls by using conserved motifs and achieves over 90% sensitivity and specificity.

PathoBERT improves bacterial pathogenicity prediction from short reads by combining a DNABERT encoder with attention-based feature extraction, achieving better performance than existing methods.

The paper identifies a metastasis-initiating-like state in prostate cancer linked to FEV, tumor plasticity, and immune niche remodeling through integrated multi-omics analysis.

The paper shows that age prediction models do not establish biological age measurement and highlights the need for assumptions beyond age prediction to validate biological age constructs.

The paper develops methods to infer organ aging, senescence, and hallmark scores from blood transcriptomes and facial photographs, showing associations with mortality.

The paper develops deepHTL, a deep learning method to detect treatment effect heterogeneity in lung transplant outcomes, finding significant variation in benefits based on recipient characteristics.

GeneSPOT is a gene-centric framework that characterizes spatial gene relationships and their reorganization across biological conditions and time.

This paper introduces a standardized framework for assessing FAIRness in agrosystem research data infrastructures using a multi-dimensional scoring system.