Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Transcriptomics
Learned models recover structure from sparse spatial and single-cell transcriptomes
Graph, contrastive and diffusion models now pair learned representations with spatial, histology and biological priors across transcriptomic datasets.
Single-cell
Computational methods decode single-cell and spatial omics across platforms and tissues
Matched multi-platform benchmarks and larger atlases now test spatial and single-cell methods against shared reference data.
Single-cell
Learned embeddings integrate single-cell, spatial and bulk omics across regulatory tasks
Pretrained multimodal models and cross-species alignment now extend single-cell and spatial representations to larger datasets and new biological questions.
Structural bio & protein design
Deep learning and energy methods extend RNA, protein, and peptide structure design
AlphaFold3-era tools for RNA, protein-RNA, and peptide structures are being benchmarked and applied to design, model quality assessment, and variant analysis.
Genomics
Statistical methods for omic data: mediation, network inference, and variable selection
Recent preprints and publications apply Bayesian, mixed, and knockoff-based models to microbiome, single-cell, and multi-omic datasets.