{"id":"W4226190998","doi":"10.3389/fmicb.2021.764058","title":"CANT-HYD: A Curated Database of Phylogeny-Derived Hidden Markov Models for Annotation of Marker Genes Involved in Hydrocarbon Degradation","year":2022,"lang":"en","type":"article","venue":"Frontiers in Microbiology","topic":"Biofuel production and bioconversion","field":"Engineering","cited_by":79,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund; Government of Alberta; Genome Canada","keywords":"Metagenomics; Degradation (telecommunications); Gene; Phylum; Computational biology; Sequence database; Genome; Hydrocarbon; Biology; Database; Computer science; Chemistry; Genetics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.001467454,0.003576409,0.002291232,0.0046351,0.001570968,0.001847218,0.002498944,0.002444211,0.01258398],"category_scores_gemma":[0.005111233,0.001703202,0.002356146,0.005120859,0.0004229106,0.001605399,0.00241313,0.002117535,0.009306539],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001398467,"about_ca_system_score_gemma":0.003213772,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00482432,"about_ca_topic_score_gemma":0.01386241,"domain_scores_codex":[0.9991515,0.0001616896,0.0001328999,0.0002951961,0.0001686102,0.00009001212],"domain_scores_gemma":[0.9980512,0.0009691582,0.0002843042,0.0003016107,0.0002260212,0.000167628],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.005053469,0.0008370017,0.03299337,0.02852781,0.002382596,0.00270647,0.001745178,0.05470419,0.2156658,0.01370351,0.4689139,0.1727669],"study_design_scores_gemma":[0.0009410604,0.0007684595,0.02900644,0.002204839,0.001463883,0.001541174,0.000684546,0.1465183,0.05779972,0.02162636,0.7368131,0.0006320506],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.03026618,0.004409956,0.08223,0.0002354583,0.0002153598,0.0003160989,0.8125017,0.06418152,0.005643672],"genre_scores_gemma":[0.03569056,0.001814274,0.08558554,0.0001452179,0.00003646104,0.0007544578,0.8696222,0.005382994,0.0009682688],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01258398,"threshold_uncertainty_score":0.04209757,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01326671482194577,"score_gpt":0.2049205143376406,"score_spread":0.1916537995156948,"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."}}