{"id":"W4220876663","doi":"10.11606/t.95.2022.tde-17022022-091454","title":"Machine learning prediction in genomic sequences of prokaryotic viruses from metagenomic datasets","year":2022,"lang":"en","type":"dissertation","venue":"","topic":"Bacteriophages and microbial interactions","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Discovery Air (Canada)","funders":"Coordenação de Aperfeiçoamento de Pessoal de Nível Superior; Fundação de Amparo à Pesquisa do Estado de São Paulo","keywords":"Metagenomics; Patience; Computational biology; Autonomy; Biology; Artificial intelligence; Computer science; Genetics; Gene; Psychology; Political science","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":[0.001675165,0.0004572863,0.0004225457,0.001179118,0.0004368207,0.001078191,0.0004289935,0.0005946037,0.0003310374],"category_scores_gemma":[0.005404505,0.0002219556,0.0008043507,0.001183534,0.0002994381,0.00111677,0.0005463273,0.001178274,0.0003163015],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005333279,"about_ca_system_score_gemma":0.0006058355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002276394,"about_ca_topic_score_gemma":0.003710743,"domain_scores_codex":[0.9993468,0.0001978148,0.00004157982,0.000241497,0.00007290835,0.00009936578],"domain_scores_gemma":[0.9965359,0.002614998,0.0002924236,0.000198576,0.0001958342,0.0001622359],"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":[0.001956765,0.001032808,0.4115774,0.000764524,0.0006675735,0.0005764005,0.0008754547,0.1924771,0.1064012,0.003315378,0.004705815,0.2756496],"study_design_scores_gemma":[0.00003526664,0.0002564033,0.1034223,0.00007527475,0.0001239155,0.0001580843,0.0005831969,0.8696789,0.015703,0.007752704,0.002169152,0.00004185753],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9840443,0.0006879849,0.01183941,0.0002386458,0.00002405399,0.00002547886,0.00238054,0.0001847364,0.0005747632],"genre_scores_gemma":[0.9696938,0.0003154579,0.01940686,0.0000461173,0.00003290205,0.00004149225,0.01007065,0.0000371355,0.0003555867],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002276394,"threshold_uncertainty_score":0.008859217,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01390560452523191,"score_gpt":0.2599355016125069,"score_spread":0.246029897087275,"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."}}