{"id":"W4402029625","doi":"10.1038/s41467-024-51891-9","title":"scCAD: Cluster decomposition-based anomaly detection for rare cell identification in single-cell expression data","year":2024,"lang":"en","type":"article","venue":"Nature Communications","topic":"Single-cell and spatial transcriptomics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Natural Science Foundation of China","keywords":"Computational biology; Cell type; Identification (biology); Cluster analysis; Cell; Computer science; Biology; Benchmark (surveying); Transcriptome; Bioinformatics; Gene expression; Gene; Artificial intelligence; Genetics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002647015,0.0001347947,0.00009627989,0.0001210308,0.0001854088,0.0001124362,0.0009456691,0.0003850225,0.000003900708],"category_scores_gemma":[0.00005839282,0.0001396691,0.00007359416,0.0001949647,0.00005254744,0.00002439335,0.0001763451,0.0003649727,0.000006558765],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004373098,"about_ca_system_score_gemma":0.00007861197,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000165373,"about_ca_topic_score_gemma":0.0008434568,"domain_scores_codex":[0.998961,0.0001143231,0.0002769531,0.0004076322,0.00009083863,0.0001492239],"domain_scores_gemma":[0.9978637,0.0001275799,0.0000660447,0.001797581,0.0001041025,0.00004105271],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00006126727,0.0002605168,0.0003152067,0.00007638933,0.000006309209,2.577753e-7,0.00003757283,0.0001299613,0.9953748,0.00001632867,0.002156181,0.001565228],"study_design_scores_gemma":[0.0006193413,0.00008119756,0.0004865052,0.00007265935,0.00003553347,0.000002015211,0.00003517456,0.02886398,0.8795575,0.00004771214,0.09000093,0.0001974082],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6836041,0.04072446,0.2648465,0.003511909,0.001771064,0.002100835,0.001390005,0.0002096296,0.001841457],"genre_scores_gemma":[0.9831911,0.0001824788,0.009242418,0.0002557644,0.0001129716,0.0000935459,0.00671688,0.00003435609,0.0001704242],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.299587,"threshold_uncertainty_score":0.569554,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02911704681770159,"score_gpt":0.3079217607430363,"score_spread":0.2788047139253347,"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."}}