{"entity":{"id":"scfoundation","kind":"technology","name":"scFoundation (BioMap)","aka":[],"tldr":"scFoundation is a 100-million-parameter model trained on 50 million cells, from China's BioMap.","summary":"scFoundation is a single-cell foundation model from China's BioMap that uses masked expression modelling over the full vocabulary of about 19k human genes, with an asymmetric encoder-decoder so the expensive encoder sees only expressed genes. The Nature Methods 2024 paper describes a 100-million-parameter model trained on 50 million cells and applies it to read-depth enhancement, drug response prediction and perturbation prediction. It is aimed at computational groups who want whole-transcriptome modelling rather than a truncated gene list, including for cancer cell line drug response. The design is compute-heavy, which limits who can fine-tune or run it at scale, and like its peers its perturbation predictions await independent benchmarking. For a newcomer: scFoundation models every gene in a cell at once and tries to predict how cells will respond to drugs.","status":"emerging","asOf":"2026-09-08","links":[{"label":"Nature Methods 2024","url":"https://doi.org/10.1038/s41592-024-02305-7"}],"tags":["foundation-model","virtual-cell"],"related":[],"cancers":[],"sections":["ai-computation"],"technologies":[],"targets":[],"drugs":[],"companies":[],"institutions":[],"pathways":[],"terms":["gene-expression","cell-line"],"trials":[],"people":[],"bottlenecks":[],"keyPapers":["paper-hao-nat-methods"],"journals":[],"dependsOn":[],"notes":[],"principle":"scFoundation uses masked expression modelling with a full gene vocabulary.","strengths":["Full-gene modelling"],"limitations":["Compute-heavy"],"since":2024},"route":"/technologies/scfoundation/","neighbours":{"section":[{"id":"ai-computation","kind":"section","name":"AI & Computation","route":"/fronts/ai-computation/"}],"term":[{"id":"cell-line","kind":"term","name":"Cell line","route":"/terms/cell-line/"},{"id":"gene-expression","kind":"term","name":"Gene expression","route":"/terms/gene-expression/"},{"id":"single-cell-foundation-models","kind":"term","name":"Single-cell and transcriptome foundation models: UCE, GeneCompass, BulkFormer, BulkRNABert","route":"/terms/single-cell-foundation-models/"}],"paper":[{"id":"paper-hao-nat-methods","kind":"paper","name":"Large-scale foundation model on single-cell transcriptomics","route":"/key-papers/paper-hao-nat-methods/"}],"technology":[{"id":"cellfm","kind":"technology","name":"CellFM","route":"/technologies/cellfm/"}],"roadmap":[{"id":"virtual-cell","kind":"roadmap","name":"Virtual cell roadmap: from bulk omics to a predictive model of a cancer cell","route":"/roadmaps/virtual-cell/"}]}}