Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
ML / AI methods
Graph and attention networks integrate single-cell and spatial multi-omics data
Graph and foundation models now infer spatial expression and cell-cell communication from histology and multi-omics measurements.
Genomics
Pangenome and graph references paired with dedicated structural variant calling
Pangenome and graph-based references are now paired with specific calling logic, with reported SV F1 gains in PanVariants and SVScope.
Genomics
Transformer and deep learning tools benchmarked against established genomics methods
Recent preprints and papers benchmark transformer and deep learning models for long-read taxonomy, gene prediction, and proteomics against established tools.
Genomics
Computational inference of microbial traits from gut, viral, and pathogen genomes
Predictive models are now validated across independent cohorts, while benchmarks scrutinize the inference methods themselves.
ML / AI methods
Benchmarks expose that larger biological AI models often fail to beat simple baselines
Recent preprints and published benchmarks test whether biological prediction models outperform additive, classical, or descriptor-based baselines.