{"id":"W2438022564","doi":"10.1101/059881","title":"BlastPhyMe: A toolkit for rapid generation and analysis of protein-coding sequence datasets","year":2016,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; GenBank; Coding (social sciences); Data mining; Graphical user interface; Replicate; Phylogenetic tree; Selection (genetic algorithm); Visualization; Artificial intelligence; Programming language; Biology; 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":[0.004492479,0.004335858,0.003545528,0.005368215,0.001639431,0.003495241,0.006050392,0.002289564,0.03430695],"category_scores_gemma":[0.008803291,0.003662299,0.002486894,0.004262052,0.0008858062,0.00433858,0.006748447,0.005816027,0.04538257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007859872,"about_ca_system_score_gemma":0.002513455,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001608053,"about_ca_topic_score_gemma":0.00208789,"domain_scores_codex":[0.9970402,0.000838205,0.0004030917,0.0005393033,0.0009433292,0.0002358143],"domain_scores_gemma":[0.9975968,0.001084273,0.0003025492,0.0004408429,0.0002960122,0.0002795799],"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":[0.001188828,0.0001634958,0.001928355,0.006384255,0.0006386648,0.001428791,0.0007751047,0.004124704,0.05241937,0.007881165,0.7511033,0.1719638],"study_design_scores_gemma":[0.0006586947,0.0001746262,0.004922119,0.00115252,0.0001987123,0.001915543,0.0002360181,0.03823151,0.032958,0.03020547,0.8887925,0.0005541978],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.003262969,0.002548016,0.4580392,0.0006300285,0.0006519132,0.000598206,0.0945261,0.434954,0.004789543],"genre_scores_gemma":[0.01293714,0.002899593,0.6649322,0.001032862,0.0002097921,0.003657161,0.2325828,0.07645049,0.005298006],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.03430695,"threshold_uncertainty_score":0.1147681,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02397416342418301,"score_gpt":0.2422565394633419,"score_spread":0.2182823760391588,"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."}}