{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002277183,0.001477094,0.001376664,0.002972876,0.001016833,0.001502399,0.001602044,0.001089314,0.001302747],"category_scores_gemma":[0.004693222,0.0004688536,0.001870837,0.002329048,0.0009650637,0.001021249,0.002378523,0.002182648,0.001321792],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008665143,"about_ca_system_score_gemma":0.002097639,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006901972,"about_ca_topic_score_gemma":0.01037868,"domain_scores_codex":[0.9982835,0.0003233392,0.0001155812,0.0005527321,0.0005643053,0.0001605346],"domain_scores_gemma":[0.9980993,0.000656787,0.0001903982,0.0003906812,0.0005074231,0.0001554692],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008006507,0.0003569395,0.02395457,0.0006861674,0.0006667875,0.0005626446,0.0007530213,0.1034073,0.1193962,0.008856528,0.03629103,0.7042681],"study_design_scores_gemma":[0.00005477385,0.00007322145,0.004381516,0.00002219662,0.00005351237,0.0003045249,0.0001768351,0.9505067,0.02367958,0.01122648,0.009446747,0.00007388349],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02345551,0.0005953763,0.9656373,0.0002851073,0.0001578607,0.0001456745,0.001363263,0.007773566,0.0005863166],"genre_scores_gemma":[0.1605609,0.0004193307,0.828142,0.0003482327,0.0001213207,0.0003755049,0.007281633,0.0007396326,0.00201126],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006901972,"threshold_uncertainty_score":0.01372361,"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."}}