{"id":"W4403090726","doi":"10.1101/2024.10.02.616248","title":"PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Single-cell and spatial transcriptomics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Wellcome Trust","keywords":"Benchmarking; Perturbation (astronomy); Foundation (evidence); Computer science; Econometrics; Mathematics; Physics; Economics; Law; Political science","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":[0.002298262,0.001461342,0.001007394,0.001241087,0.000540701,0.001208119,0.002056887,0.002496439,0.003147678],"category_scores_gemma":[0.008840717,0.0004850002,0.001527974,0.0009055146,0.0007890497,0.001447649,0.001261149,0.001574037,0.001264236],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001285162,"about_ca_system_score_gemma":0.001504365,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008815541,"about_ca_topic_score_gemma":0.01003967,"domain_scores_codex":[0.9993694,0.0002242448,0.00003915591,0.0001747725,0.0001407503,0.00005165687],"domain_scores_gemma":[0.9972479,0.001800656,0.0001395371,0.0004058146,0.0002957775,0.0001103079],"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":[0.0001654427,0.00009158962,0.003084448,0.0003348115,0.0001489499,0.00008943911,0.00005158061,0.9647849,0.002264762,0.003842644,0.006625638,0.01851582],"study_design_scores_gemma":[0.00001698151,0.00004476558,0.0002401295,0.00001444907,0.000008753845,0.00001883602,0.00001527963,0.9941573,0.001227391,0.003267682,0.0009778555,0.00001054138],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4130078,0.00358005,0.5053299,0.001879106,0.0005649469,0.0003440279,0.02650323,0.03824621,0.01054489],"genre_scores_gemma":[0.8051894,0.0007732787,0.1593932,0.0005895176,0.00008322077,0.0005284887,0.02840373,0.002617359,0.002421811],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008815541,"threshold_uncertainty_score":0.01752841,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01549444564827791,"score_gpt":0.2126725885666566,"score_spread":0.1971781429183787,"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."}}