Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
ML / AI methods
Supervised and self-supervised models for protein prediction and evaluation design
Recent preprints report new protein models alongside evaluation fixes, including default-threshold and shortcut-bias audits in EC and PPI prediction.
ML / AI methods
ML models infer RNA-binding and transcription factor regulatory interactions
Recent preprints and publications apply foundation and interpretable models to UTR, DNA, RBP-splicing, and TF regulatory data.
Structural bio & protein design
Interface chemistry and heterogeneity set transport rates across membranes and matrices
Mesoscale and multiscale simulations now quantify how local interface chemistry couples to ion, proton, and nanoparticle transport outcomes.
Single-cell
Spatial and single-cell transcriptomic tools are built and independently benchmarked
Matched spatial and single-nucleus profiling from one section, plus standardized pipelines, expand cell-level information from limited material.
Transcriptomics
Learned models infer gene expression from degraded, spatial and histology-derived inputs
Recent preprints and one published study extend learned models to degraded FFPE, spatial and histology-linked transcriptomic data.