{"id":"W2170330565","doi":"10.1093/bioinformatics/btr353","title":"A tree-based approach for motif discovery and sequence classification","year":2011,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Princess Margaret Cancer Centre; University Health Network; University of Toronto; Ontario Institute for Cancer Research","funders":"Canadian Institutes of Health Research","keywords":"Computer science; Multiple sequence alignment; Artificial intelligence; Pattern recognition (psychology); Parallelizable manifold; Sequence (biology); Tree (set theory); Data mining; Sequence alignment; Algorithm; Mathematics; Biology; Peptide sequence; Genetics","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":[0.0035265,0.001864311,0.001914823,0.005665063,0.001376419,0.002785565,0.004597923,0.00264324,0.005059199],"category_scores_gemma":[0.01442382,0.0009338061,0.003076132,0.008272955,0.00115838,0.003347519,0.002154122,0.003521499,0.003873089],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001580932,"about_ca_system_score_gemma":0.002513343,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004401864,"about_ca_topic_score_gemma":0.005013508,"domain_scores_codex":[0.9953706,0.001060173,0.000488913,0.001383606,0.001478984,0.0002178709],"domain_scores_gemma":[0.9927884,0.003947342,0.0005399,0.001043906,0.001419681,0.0002607397],"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":[0.0003538449,0.0003678231,0.00497186,0.0007925049,0.0004039681,0.0002652761,0.0002729436,0.09279051,0.01298134,0.02474659,0.02490295,0.8371504],"study_design_scores_gemma":[0.00005317733,0.00009632009,0.0005979054,0.00005150994,0.0000476649,0.0002928368,0.00004663558,0.9302871,0.003664678,0.05407399,0.01076064,0.00002754744],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002402987,0.0002527939,0.9930074,0.0001598187,0.00004368492,0.0002290315,0.0008250337,0.002652474,0.0004266163],"genre_scores_gemma":[0.01859599,0.0001737593,0.9766175,0.0001319389,0.00004583105,0.0004545781,0.003038551,0.0002168725,0.0007250641],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005665063,"threshold_uncertainty_score":0.01865011,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03837618621678023,"score_gpt":0.2335725413216516,"score_spread":0.1951963551048713,"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."}}