{"id":"W4410399785","doi":"10.1016/j.ajhg.2025.04.010","title":"A flexible machine learning Mendelian randomization estimator applied to predict the safety and efficacy of sclerostin inhibition","year":2025,"lang":"en","type":"article","venue":"The American Journal of Human Genetics","topic":"Genetic Associations and Epidemiology","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; Institut universitaire de cardiologie et de pneumologie de Québec; Université de Montréal","funders":"Canadian Institutes of Health Research; Fonds de Recherche du Québec - Santé; Institut de Valorisation des Données","keywords":"Mendelian randomization; Sclerostin; Estimator; Randomization; Computer science; Machine learning; Artificial intelligence; Medicine; Mathematics; Statistics; Internal medicine; Biology; Randomized controlled trial; Genetics; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02540872,0.0006116658,0.002081038,0.001145076,0.0004043283,0.0008910503,0.001476685,0.001511239,0.002301533],"category_scores_gemma":[0.05837003,0.0005043554,0.001355258,0.00071281,0.0008755925,0.0007992983,0.001013703,0.001862651,0.0003717678],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004723366,"about_ca_system_score_gemma":0.001214798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001764392,"about_ca_topic_score_gemma":0.001593525,"domain_scores_codex":[0.9890375,0.008925643,0.0003707809,0.0009877078,0.000454095,0.0002243408],"domain_scores_gemma":[0.9351865,0.05759162,0.002646175,0.002785131,0.001364587,0.000426001],"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":[0.006766937,0.000775434,0.09768728,0.0003839226,0.002996977,0.0006439413,0.0002097078,0.3980954,0.005138766,0.03611819,0.005044619,0.4461388],"study_design_scores_gemma":[0.0002944158,0.0003703277,0.00452739,0.00002650013,0.0001558453,0.0001689234,0.000007706052,0.9852448,0.0005135012,0.008141011,0.000526139,0.00002336905],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1140459,0.0006137554,0.8829811,0.0004589225,0.0001103183,0.0002059838,0.0003255365,0.0008401932,0.0004183266],"genre_scores_gemma":[0.7933939,0.0002508709,0.2032729,0.0003716695,0.0001315723,0.0004216663,0.0007512564,0.00008961745,0.001316597],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02540872,"threshold_uncertainty_score":0.1343758,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009071578261835533,"score_gpt":0.2704850063026728,"score_spread":0.2614134280408373,"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."}}