{"id":"W3008299510","doi":"10.1101/2020.02.03.932350","title":"Machine learning using intrinsic genomic signatures for rapid classification of novel pathogens: COVID-19 case study","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"SARS-CoV-2 and COVID-19 Research","field":"Medicine","cited_by":87,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Western University","funders":"","keywords":"Coronavirus disease 2019 (COVID-19); Genome; Decision tree; Artificial intelligence; Machine learning; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); Coronavirus; Computational biology; 2019-20 coronavirus outbreak; Outbreak; False positive rate; Spearman's rank correlation coefficient; Biology; Computer science; Virology; Medicine; Genetics; Gene","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007284033,0.0005551515,0.0004415761,0.001536766,0.0007153746,0.0009929745,0.0008449982,0.001428988,0.000731755],"category_scores_gemma":[0.00247126,0.0002067725,0.0004652911,0.0008460447,0.0005328326,0.0005807362,0.0006485549,0.0008952794,0.0004237585],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004658501,"about_ca_system_score_gemma":0.0003241066,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002651801,"about_ca_topic_score_gemma":0.003464999,"domain_scores_codex":[0.999517,0.00009815725,0.00005357461,0.0001103433,0.0001522794,0.00006862565],"domain_scores_gemma":[0.9990324,0.0003597907,0.0001245775,0.0001106742,0.0002109321,0.0001616963],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001139753,0.001769249,0.5400146,0.0005471979,0.000191411,0.156653,0.001979746,0.02496593,0.05471431,0.002586492,0.01637393,0.1990645],"study_design_scores_gemma":[0.0001301793,0.001283382,0.1667668,0.0001770577,0.0001577232,0.1782355,0.003859261,0.5652227,0.06190047,0.004563516,0.01752817,0.0001752228],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9724995,0.0005728381,0.02276,0.001056481,0.00008565629,0.0001428406,0.0007064919,0.0001744938,0.002001706],"genre_scores_gemma":[0.9671608,0.0003265328,0.03038977,0.0001509161,0.0001012279,0.0000298442,0.001047316,0.00003261496,0.0007609327],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002651801,"threshold_uncertainty_score":0.005272746,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09905360487178151,"score_gpt":0.3402559393750239,"score_spread":0.2412023345032424,"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."}}