{"id":"W4307522818","doi":"10.1101/2022.10.21.22281368","title":"A practical approach to curate clonal hematopoiesis of indeterminate potential in human genetic datasets","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Acute Myeloid Leukemia Research","field":"Medicine","cited_by":58,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; St. Joseph’s Healthcare Hamilton; Impact; Queen's University","funders":"National Institute of Diabetes and Digestive and Kidney Diseases; Medical Research Council; Vanderbilt University; Canadian Institutes of Health Research; Leukemia and Lymphoma Society; American Society of Hematology; Vanderbilt University Medical Center; Alexander and Margaret Stewart Trust; Ludwig Center at Harvard; National Institutes of Health; Pew Charitable Trusts","keywords":"Exome; Exome sequencing; Biology; Biobank; Population; Germline; Genetics; Computational biology; Mutation; Medicine; Gene","routes":{"ca_aff":true,"ca_fund":true,"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","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00119578,0.0003882815,0.0009895269,0.0008003615,0.00008080464,0.00003693198,0.000708728,0.0003385216,0.0009229444],"category_scores_gemma":[0.0005157967,0.0003785092,0.0002000641,0.0005068941,0.0001868706,0.00005788537,0.003666619,0.002517199,0.00005232123],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005213462,"about_ca_system_score_gemma":0.001059966,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004895792,"about_ca_topic_score_gemma":0.00001899948,"domain_scores_codex":[0.9956496,0.0004877233,0.0008954885,0.001039916,0.001307819,0.0006194708],"domain_scores_gemma":[0.9976413,0.0001140415,0.0002790571,0.001484105,0.0001164701,0.0003650358],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.01027439,0.01903771,0.4506354,0.03280717,0.003546977,0.05314046,0.009175294,0.01017676,0.2271425,0.003330376,0.1721027,0.008630227],"study_design_scores_gemma":[0.01242566,0.002227918,0.8878986,0.001539734,0.001404736,0.003754245,0.001748566,0.03292166,0.02430137,0.002000938,0.02679576,0.002980834],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9936652,0.00009010281,0.001127071,0.001273324,0.0001672538,0.001621593,0.0006454266,0.00004105196,0.00136903],"genre_scores_gemma":[0.9765339,0.00004188643,0.01983384,0.0002322949,0.0001926269,0.0005416933,0.002179211,0.00008938879,0.0003551083],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4372632,"threshold_uncertainty_score":0.9999903,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05718008216702608,"score_gpt":0.3802776397486847,"score_spread":0.3230975575816586,"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."}}