{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001543406,0.00067176,0.001188477,0.0007431807,0.0003199225,0.000125936,0.0004886359,0.000632374,0.00001230158],"category_scores_gemma":[0.002679553,0.0006960064,0.0003518624,0.0008550663,0.0001820993,0.00009147127,0.000620185,0.001653737,0.000007082904],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009650029,"about_ca_system_score_gemma":0.003567455,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006524533,"about_ca_topic_score_gemma":0.00001581767,"domain_scores_codex":[0.9958908,0.0002951011,0.001013312,0.001492944,0.0006848106,0.0006230535],"domain_scores_gemma":[0.9963001,0.000384806,0.0008049904,0.001268716,0.0009289098,0.0003124859],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0006658894,0.0008768159,0.03651615,0.002310967,0.0005360455,0.001314793,0.0001516854,0.0001423725,0.9573587,0.00008475707,0.0000231696,0.00001861893],"study_design_scores_gemma":[0.02800908,0.00394797,0.08011684,0.001295776,0.004478351,0.00001236659,0.001021555,0.1800287,0.6528494,0.000007003759,0.04424954,0.003983413],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9772099,0.002724312,0.01435292,0.0002215024,0.0003839667,0.004350333,0.0004253717,0.0003289861,0.000002727343],"genre_scores_gemma":[0.9892405,0.00003087762,0.005759009,0.003729122,0.000608145,0.0004054999,0.00000158598,0.0002247158,5.238911e-7],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3045093,"threshold_uncertainty_score":0.9995491,"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."}}