Top 40
View all ›PARTIAL Some sources or summaries were unavailable for this run.
BACKFILLED Code and industry signals are partial for backfilled days.
ML / AI methods
Joint latent spaces link histology, omics and single-cell data across modalities
Deep generative and graph models now map paired and unpaired multimodal measurements into shared latent spaces for downstream biological inference.
Single-cell
LLMs and network medicine are prioritizing drug pairs and Alzheimer's disease targets
Transcriptomics
Spatial omics methods couple histology with molecular profiles at subcellular resolution
Multiple new tools integrate histology or morphology with spatial transcriptomic and proteomic measurements across platforms and tissue scales.
ML / AI methods
Genomic and protein language models add evolutionary and biophysical structure
Recent preprints test whether injected structure, quantization, and fitness-fusion harnesses improve sequence-level prediction beyond scale alone.
Structural bio & protein design
Geometric and topological structure features drive binding affinity prediction models
Persistent topology, local Laplacian descriptors, and predicted geometries now feed learned models for binding affinity and complex model selection.