Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Single-cell
Computational inference of niche-level cell communication from spatial transcriptomics
Integrated resources now span over ten million spatial transcriptomes, shifting inference toward niche- and condition-defined interactions.
Structural bio & protein design
Physics-based and generative models resolve protein conformational dynamics and design
Generative, MD-based, and FRET-scale methods now resolve conformational, allosteric, and sequence-level structural features at scale.
ML / AI methods
Learned models infer tumor expression and phenotypes from histology, MRI and multiomics
Histology, MRI and multi-assay RNA data now feed deep models that predict spatial expression, immune signatures and survival-linked features.
Structural bio & protein design
Generative models expand protein sequence and structure design landscapes
Recent preprints and papers benchmark generative and language-model protein design against simple baselines and atomic-level diffusion methods.
Epigenomics
Learned models link noncoding regulatory elements and variants to genes and disease
Expanding multi-omic and 3D-informed datasets now let learned models resolve enhancer-gene links, regulatory elements and variant effects in context.