{"id":"W4392168151","doi":"10.1038/s41592-024-02201-0","title":"scGPT: toward building a foundation model for single-cell multi-omics using generative AI","year":2024,"lang":"en","type":"article","venue":"Nature Methods","topic":"Single-cell and spatial transcriptomics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1044,"is_retracted":false,"has_abstract":false,"ca_institutions":"Vector Institute; University of Toronto; University Health Network","funders":"Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; University Health Network","keywords":"Generative grammar; Inference; Synthetic biology; Computer science; Systems biology; Computational biology; Transformer; Annotation; Generative model; Artificial intelligence; Language model; Biology; Machine learning; Engineering","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001535556,0.0008596919,0.001210137,0.0009366241,0.0008893895,0.002178551,0.003343306,0.001862006,0.00636278],"category_scores_gemma":[0.00584805,0.0008944013,0.002464365,0.001008214,0.001648649,0.00211907,0.003308466,0.003202987,0.002238861],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001140742,"about_ca_system_score_gemma":0.001981226,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006471152,"about_ca_topic_score_gemma":0.00777853,"domain_scores_codex":[0.9994293,0.0002055767,0.00002689758,0.0001121115,0.0001730238,0.00005308733],"domain_scores_gemma":[0.9975674,0.001546847,0.00007590423,0.0004385191,0.0002447015,0.0001266085],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005170177,0.00003605337,0.0008823352,0.0001927908,0.0001200472,0.0001783678,0.0001774414,0.6734373,0.002732859,0.283755,0.004866521,0.03356963],"study_design_scores_gemma":[0.000005474394,0.000004484018,0.00003244373,0.00001071556,0.000008214073,0.00002013106,0.000008818249,0.9069035,0.0003947103,0.08984464,0.002761587,0.0000052994],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001663467,0.00008600772,0.9942459,0.0002297306,0.00003788624,0.00002376007,0.0002653141,0.0009506735,0.002497299],"genre_scores_gemma":[0.1823325,0.00061228,0.8066639,0.0006287671,0.0001099267,0.0004888678,0.001840543,0.001835818,0.005487286],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006471152,"threshold_uncertainty_score":0.02128559,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08220096878827339,"score_gpt":0.4072693791762135,"score_spread":0.3250684103879401,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}