{"id":"W3047701039","doi":"10.1101/2020.08.07.241729","title":"Petabase-scale sequence alignment catalyses viral discovery","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Bacteriophages and microbial interactions","field":"Environmental Science","cited_by":57,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Klaus Tschira Stiftung; University of British Columbia; Russian Science Foundation; Ministerio de Economía y Competitividad; Saint Petersburg State University; Agence Nationale de la Recherche","keywords":"Biology; Computational biology; Sequence (biology); RNA polymerase; RNA; Polymerase; RNA virus; Virology; Nucleic acid; Anticipation (artificial intelligence); Gene; Genetics; Computer science; Artificial intelligence","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":[0.003311017,0.001376302,0.001390276,0.002443622,0.001518906,0.0031043,0.001786229,0.0008973783,0.006100736],"category_scores_gemma":[0.01100732,0.0009887066,0.001176723,0.003675373,0.001104807,0.002034027,0.002782307,0.002761446,0.00738947],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007083953,"about_ca_system_score_gemma":0.001389586,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001587995,"about_ca_topic_score_gemma":0.002045522,"domain_scores_codex":[0.9975021,0.0007191377,0.0002025533,0.0007672089,0.0006600872,0.0001489519],"domain_scores_gemma":[0.9954008,0.001990514,0.000549632,0.001285263,0.0004950092,0.0002787052],"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":[0.00419106,0.0008598426,0.01684466,0.002007138,0.0009748368,0.001464457,0.001123469,0.01580235,0.5223724,0.03132305,0.1024681,0.3005688],"study_design_scores_gemma":[0.0006476003,0.0008430962,0.01395397,0.0002216355,0.0002213843,0.002277077,0.0005105489,0.2652601,0.4156821,0.04700968,0.2530472,0.0003256732],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1578947,0.003850569,0.6960707,0.002419324,0.001104618,0.00160601,0.03757294,0.08561147,0.01386968],"genre_scores_gemma":[0.1600875,0.001126584,0.7850684,0.0004136843,0.0002227894,0.0009040696,0.04516925,0.003510773,0.003496894],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006100736,"threshold_uncertainty_score":0.02040905,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01987457877817397,"score_gpt":0.235009707410126,"score_spread":0.2151351286319521,"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."}}