{"id":"W4404018000","doi":"10.1101/2024.10.30.621013","title":"MethylGPT: a foundation model for the DNA methylome","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Epigenetics and DNA Methylation","field":"Biochemistry, Genetics and Molecular Biology","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto; University Health Network","funders":"Medical Research Council; Wellcome Trust; Royal College of Physicians of Edinburgh; Scottish Funding Council; Scottish Government; Chief Scientist Office, Scottish Government Health and Social Care Directorate; National Alliance for Research on Schizophrenia and Depression","keywords":"Foundation (evidence); DNA methylation; Computer science; Computational biology; Biology; Genetics; Political science; Gene; Law; Gene expression","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001184446,0.0004679167,0.0003305825,0.0001246851,0.0002099371,0.0002854289,0.0005456524,0.0006445469,0.000009808593],"category_scores_gemma":[0.0003600575,0.0003993915,0.0003279332,0.0001843361,0.0001009536,0.000004274882,0.0007214126,0.0004378668,0.00003248416],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008586175,"about_ca_system_score_gemma":0.0006032027,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001845917,"about_ca_topic_score_gemma":0.000009007773,"domain_scores_codex":[0.9977087,0.00009547588,0.0004833246,0.001016049,0.000274924,0.000421498],"domain_scores_gemma":[0.997731,0.00008116387,0.000263963,0.00131201,0.000479311,0.0001325162],"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":[0.00003493167,0.00003685083,0.00004727303,0.0002459859,0.0002835228,0.000001163706,0.000009373251,0.003740272,0.9941053,0.001068024,0.0003989017,0.00002835363],"study_design_scores_gemma":[0.0003217771,0.00008487594,0.0008390237,0.00008304524,0.0003858222,3.450405e-9,0.000002267482,0.06612342,0.9018202,0.0002892201,0.02940791,0.0006424329],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6677015,0.02081252,0.3054188,0.0009482899,0.002418759,0.002033536,0.0004693219,0.0001629238,0.00003428601],"genre_scores_gemma":[0.9842284,0.001437348,0.01199872,0.0001776516,0.00110493,0.0007723876,0.00000516646,0.0001749015,0.0001004943],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3165269,"threshold_uncertainty_score":0.9998458,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02216721304951737,"score_gpt":0.2636990197036043,"score_spread":0.2415318066540869,"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."}}