{"id":"W2935957056","doi":"10.1016/bs.mie.2019.03.015","title":"Exploring the sequence, function, and evolutionary space of protein superfamilies using sequence similarity networks and phylogenetic reconstructions","year":2019,"lang":"en","type":"article","venue":"Methods in enzymology on CD-ROM/Methods in enzymology","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":27,"is_retracted":false,"has_abstract":false,"ca_institutions":"Canada's Michael Smith Genome Sciences Centre; Michael Smith Health Research BC; University of British Columbia","funders":"","keywords":"Phylogenetic tree; Computational biology; Biology; Sequence (biology); Protein sequencing; Similarity (geometry); Alignment-free sequence analysis; Function (biology); Phylogenetics; Sequence alignment; Evolutionary biology; Protein function; Sequence space; Protein superfamily; Genetics; Peptide sequence; Gene; Artificial intelligence; Computer 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.003089441,0.0004082217,0.0007315093,0.0002973081,0.0001927708,0.0000131705,0.0003130702,0.0004713777,0.00001822696],"category_scores_gemma":[0.0007576782,0.0003683572,0.00009274552,0.0004074946,0.001444065,0.000009335161,0.0004572439,0.0006112689,7.95134e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006487168,"about_ca_system_score_gemma":0.0001213245,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002706893,"about_ca_topic_score_gemma":0.0002878844,"domain_scores_codex":[0.9938434,0.00374562,0.0007211593,0.0009834709,0.0001007759,0.0006056082],"domain_scores_gemma":[0.9978423,0.001017892,0.0002514052,0.0006905741,0.0001207844,0.00007701004],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0002133218,0.00006681813,0.09634879,0.0000492648,0.00009728138,0.000005617138,0.0001969989,0.008197581,0.8793303,0.002482773,0.000006174571,0.01300503],"study_design_scores_gemma":[0.001778756,0.001868032,0.9142321,0.0001208882,0.0001123492,0.0007125487,0.001455955,0.002396197,0.04372416,0.03083423,0.001879782,0.0008850635],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9777738,0.005880902,0.0142682,0.0001828155,0.0009504728,0.000713484,0.00001906539,0.000008428255,0.0002028504],"genre_scores_gemma":[0.6149396,0.001122866,0.3835067,0.0001323269,0.00009109218,0.0001276009,0.000005419605,0.00002813884,0.00004626745],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8356062,"threshold_uncertainty_score":0.9998769,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1068666795368128,"score_gpt":0.3591240988197934,"score_spread":0.2522574192829806,"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."}}