{"id":"W4410947908","doi":"10.1101/2025.05.29.656931","title":"Deviation Error: assessing predictions for replicate measurements in genomics and beyond","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Replicate; Artificial intelligence; Machine learning; Computer science; Econometrics; Data science; Statistics; Mathematics","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":[0.04099003,0.001846825,0.001572139,0.003343628,0.0009182181,0.003087218,0.002216055,0.003363291,0.001089553],"category_scores_gemma":[0.1159984,0.000560337,0.001541632,0.003142352,0.003090225,0.003405011,0.002928697,0.003452034,0.0006686173],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001953929,"about_ca_system_score_gemma":0.001486776,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002702462,"about_ca_topic_score_gemma":0.00188779,"domain_scores_codex":[0.9802065,0.008023665,0.001394079,0.004432506,0.005418745,0.0005244573],"domain_scores_gemma":[0.8901238,0.07919329,0.007060387,0.01584083,0.006642387,0.001139328],"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":[0.001421206,0.0004233229,0.1473765,0.0007798037,0.001401464,0.0003895811,0.0005839764,0.7058994,0.01739079,0.01644745,0.007178952,0.1007076],"study_design_scores_gemma":[0.00004338391,0.0002858248,0.02834938,0.0001114766,0.00009066472,0.0002858531,0.0001743849,0.9139205,0.01979492,0.03392753,0.002893751,0.0001224253],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3084606,0.001871937,0.6767983,0.001361412,0.0003136068,0.0002074173,0.004011371,0.004256663,0.00271872],"genre_scores_gemma":[0.8861173,0.0002350706,0.1049373,0.0004930209,0.00007777399,0.0002228616,0.006294388,0.0007197454,0.0009026576],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04099003,"threshold_uncertainty_score":0.2167786,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03470505324484557,"score_gpt":0.2802541004175138,"score_spread":0.2455490471726682,"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."}}