{"id":"W3171310600","doi":"10.3389/fgene.2021.647436","title":"A Machine Learning Method to Identify Genetic Variants Potentially Associated With Alzheimer’s Disease","year":2021,"lang":"en","type":"article","venue":"Frontiers in Genetics","topic":"Genetic Associations and Epidemiology","field":"Biochemistry, Genetics and Molecular Biology","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Neurological Disorders and Stroke; National Heart, Lung, and Blood Institute; National Institutes of Health; Karl-Franzens-Universität Graz; Texas Alzheimer's Research and Care Consortium; Österreichische Forschungsförderungsgesellschaft; National Institute on Aging; Medizinische Universität Graz; Oesterreichische Nationalbank; Case Western Reserve University; University of Toronto; European Commission; Austrian Science Fund; National Institute on Deafness and Other Communication Disorders; University of Miami; EU Joint Programme – Neurodegenerative Disease Research; National Human Genome Research Institute; Nederlandse Organisatie voor Wetenschappelijk Onderzoek; Vanderbilt University","keywords":"Overfitting; Neuropathology; Single-nucleotide polymorphism; Disease; Exome; Genetic association; Computational biology; Bioinformatics; Alzheimer's disease; Biology; Exome sequencing; Medicine; Genetics; Artificial intelligence; Gene; Genotype; Artificial neural network; Computer science; Internal medicine; Mutation","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005090671,0.0002601992,0.0003576416,0.00011534,0.000128758,0.00004294168,0.000261794,0.0002257106,0.00003028146],"category_scores_gemma":[0.0006047268,0.000271368,0.0001076084,0.0003945097,0.00004809785,0.000002969912,0.0002209769,0.0002077381,0.000008742036],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004679704,"about_ca_system_score_gemma":0.0002692365,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002708814,"about_ca_topic_score_gemma":0.000143978,"domain_scores_codex":[0.9974276,0.0006849916,0.0004184571,0.0006950046,0.0002237083,0.0005502501],"domain_scores_gemma":[0.9988613,0.00003243393,0.0001597213,0.0004430854,0.000218724,0.0002846964],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001122525,0.0001640571,0.9049457,0.000007650669,0.0004385836,0.000114528,0.00009439606,0.06421423,0.009187078,0.000007043841,0.00445302,0.0162615],"study_design_scores_gemma":[0.001040132,0.0002559346,0.9724213,0.0000311204,0.0002502899,0.00001575943,0.0001344327,0.008986519,0.001405338,0.0003831019,0.01461356,0.0004624475],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4685151,0.01282815,0.5166096,0.0006286947,0.0005853542,0.0003752142,0.00008039384,0.00002563348,0.0003518611],"genre_scores_gemma":[0.6415442,0.0008220207,0.3549582,0.0007786333,0.0001346574,0.00004903347,0.0005277584,0.00006232754,0.001123155],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.1730292,"threshold_uncertainty_score":0.9999738,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01550427350432422,"score_gpt":0.2937245734043784,"score_spread":0.2782202999000541,"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."}}