{"id":"W1996719109","doi":"10.1371/journal.pone.0039475","title":"SECOM: A Novel Hash Seed and Community Detection Based-Approach for Genome-Scale Protein Domain Identification","year":2012,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"King Abdullah University of Science and Technology","keywords":"Genome; Computational biology; Biology; Proteome; Hash function; Protein domain; Computer science; Genetics; 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":[0.0008474769,0.001728544,0.00139264,0.003578296,0.0008213632,0.0008131856,0.001876178,0.001204868,0.001515292],"category_scores_gemma":[0.001843053,0.0006632964,0.001523442,0.002222614,0.0006133234,0.002056067,0.002215369,0.001059648,0.00110766],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000507021,"about_ca_system_score_gemma":0.0009844538,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002409482,"about_ca_topic_score_gemma":0.003919618,"domain_scores_codex":[0.9991503,0.0001338593,0.00004642763,0.000270569,0.0003058688,0.00009300561],"domain_scores_gemma":[0.999078,0.0002835372,0.0001313104,0.0001711951,0.0002157702,0.0001200051],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00105237,0.0005646186,0.01705128,0.0008168536,0.0005223841,0.001202368,0.0004496507,0.05893614,0.1606609,0.01334519,0.01609634,0.7293019],"study_design_scores_gemma":[0.00007001502,0.0001503354,0.002278931,0.00002326191,0.00005193794,0.0005879991,0.0001530002,0.9478149,0.02787489,0.01410179,0.006836887,0.00005602436],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02751125,0.0004224129,0.9630052,0.0001325622,0.00005129144,0.0002164979,0.0007594926,0.007090552,0.0008107417],"genre_scores_gemma":[0.1301717,0.0002292216,0.8643942,0.0001223305,0.00003926852,0.0001878796,0.002824758,0.0003280515,0.001702553],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003578296,"threshold_uncertainty_score":0.005069196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04155282900149822,"score_gpt":0.2207617751314493,"score_spread":0.179208946129951,"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."}}