{"id":"W2887325828","doi":"10.1109/tnb.2018.2845741","title":"Discovering Patterns From Sequences Using Pattern-Directed Aligned Pattern Clustering","year":2018,"lang":"en","type":"article","venue":"IEEE Transactions on NanoBioscience","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Cluster analysis; Leverage (statistics); Computer science; Computational biology; Breakpoint; Data mining; Pattern recognition (psychology); Algorithm; Artificial intelligence; Genetics; Biology","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007326123,0.001292047,0.001117272,0.006066431,0.000732159,0.001038453,0.001438567,0.0009250278,0.001492709],"category_scores_gemma":[0.003209252,0.0005269852,0.001230426,0.004692407,0.0006666277,0.001304842,0.001146873,0.001000642,0.001427882],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004975629,"about_ca_system_score_gemma":0.001424095,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003067835,"about_ca_topic_score_gemma":0.003857707,"domain_scores_codex":[0.9987541,0.0001573423,0.0001465445,0.0005022163,0.0003664397,0.0000732261],"domain_scores_gemma":[0.9988852,0.0003777666,0.0001801417,0.0002023758,0.0003028533,0.00005162505],"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":[0.0002671567,0.0003460266,0.006017149,0.0008447595,0.0002951332,0.0003912706,0.0005019049,0.05623886,0.04975943,0.0068944,0.006181409,0.8722624],"study_design_scores_gemma":[0.00005903595,0.0001579804,0.002949514,0.00006079297,0.00009529297,0.0004678973,0.000341569,0.9288494,0.02406288,0.03265156,0.01024723,0.00005680433],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02664852,0.0003389868,0.9676413,0.0001488938,0.0000319366,0.0003337653,0.0009296981,0.003037299,0.0008895369],"genre_scores_gemma":[0.09167792,0.0003099511,0.9029405,0.00007696956,0.00002748258,0.0003957195,0.003481135,0.0001873062,0.0009029997],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006066431,"threshold_uncertainty_score":0.006099939,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0211234114176969,"score_gpt":0.2558546205686612,"score_spread":0.2347312091509643,"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."}}