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 autoencoder models now target domain identification, integration and deconvolution across spatial and single-cell data.
ML / AI methods
Machine learning infers spatial, perturbation and multi-omics biology from omics data
Transformer and graph models now predict perturbation responses, spatial expression from histology and cancer subtypes.
ML / AI methods
Protein language models augmented with structure, dynamics and synthetic sequences
Pretrained protein language models are being coupled with molecular dynamics, structural descriptors, and synthetic training data for function prediction.
Transcriptomics
Methods infer cell-cell communication and gene co-expression in single-cell data
Single-cell and spatial resolution now allows intercellular interactions and gene programs to be inferred in situ across tissues and experimental conditions.
Single-cell
Benchmarks and diagnostics for single-cell and bulk transcriptomics inference methods
Recent published benchmarks and diagnostic frameworks test when single-cell inference methods, including foundation models, perform reliably under known…