{"id":"W3203061596","doi":"10.1101/2021.09.28.462270","title":"Data-driven approaches for genetic characterization of SARS-CoV-2 lineages","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"SARS-CoV-2 and COVID-19 Research","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre Hospitalier Universitaire Sainte-Justine; Mila - Quebec Artificial Intelligence Institute; McGill University; Université de Montréal; Montreal Heart Institute","funders":"Fonds de Recherche du Québec - Santé; Canadian Institutes of Health Research; Natural Sciences and Engineering Research Council of Canada; Institut de Cardiologie de Montréal; Institut de Valorisation des Données; Compute Canada; Fondation Institut de Cardiologie de Montréal; National Science Foundation","keywords":"Context (archaeology); Biology; Genomics; Population; Genome; Computational biology; Phylogenomics; Pandemic; Phylogenetic tree; Genetics; Evolutionary biology; Infectious disease (medical specialty); Coronavirus disease 2019 (COVID-19); Disease; Gene; Clade; Medicine","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":[0.004813506,0.0007112371,0.0009417202,0.00389267,0.0006100389,0.001438282,0.001809664,0.001398925,0.001261019],"category_scores_gemma":[0.01987682,0.000653804,0.001754099,0.002188674,0.0008788753,0.00113444,0.001906234,0.001436143,0.0002410882],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001233231,"about_ca_system_score_gemma":0.001587007,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005074745,"about_ca_topic_score_gemma":0.004559551,"domain_scores_codex":[0.9982065,0.0009198937,0.0001315489,0.0003228999,0.0003396198,0.00007961887],"domain_scores_gemma":[0.9892271,0.008133105,0.0008129948,0.0005917891,0.0009516082,0.0002834535],"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":[0.00007633306,0.00009480037,0.01144637,0.000112406,0.000203387,0.0001778273,0.0001120277,0.9487039,0.001601646,0.01567196,0.0006114173,0.02118786],"study_design_scores_gemma":[0.000004635683,0.0000084373,0.0003263151,0.000006697561,0.000006950203,0.000009621136,0.00001826286,0.9911832,0.0001647721,0.008000685,0.0002661379,0.000004375986],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05510876,0.0003066451,0.9409368,0.0006426211,0.00005212476,0.0001478742,0.001637383,0.0005261227,0.0006416025],"genre_scores_gemma":[0.4946154,0.0002635364,0.4998595,0.0002661602,0.00009239694,0.0005262143,0.003726773,0.0001418879,0.0005081382],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005074745,"threshold_uncertainty_score":0.02545655,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1092088375131592,"score_gpt":0.3204641446402671,"score_spread":0.211255307127108,"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."}}