{"id":"W2089818979","doi":"10.1109/tcbb.2011.142","title":"Clustering 100,000 Protein Structure Decoys in Minutes","year":2011,"lang":"en","type":"article","venue":"IEEE/ACM Transactions on Computational Biology and Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Cluster analysis; Pairwise comparison; Computer science; Cluster (spacecraft); Centroid; Benchmark (surveying); Pruning; Data mining; Hierarchical clustering; Pattern recognition (psychology); Single-linkage clustering; Selection (genetic algorithm); Artificial intelligence; Algorithm; Correlation clustering; CURE data clustering algorithm","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":[0.001109965,0.001627334,0.001693549,0.001287169,0.0009117676,0.001029102,0.001924827,0.0006008848,0.006640034],"category_scores_gemma":[0.004483297,0.0005616337,0.0008021791,0.003068881,0.0003491352,0.0008862549,0.0008455649,0.0006880977,0.004311729],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009222364,"about_ca_system_score_gemma":0.002110495,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01544665,"about_ca_topic_score_gemma":0.0163924,"domain_scores_codex":[0.9983805,0.0002116165,0.0001386307,0.0004846447,0.0005803423,0.0002041787],"domain_scores_gemma":[0.9984004,0.0004192192,0.00009200921,0.0003572707,0.0005463709,0.0001846473],"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":[0.003939144,0.0008897113,0.01453998,0.001386939,0.0007466838,0.0005070834,0.001009339,0.1164463,0.08629079,0.003359036,0.1811077,0.5897772],"study_design_scores_gemma":[0.000856142,0.0007963271,0.02598915,0.00009077448,0.0002770936,0.0004854508,0.001220512,0.779319,0.09577017,0.00966306,0.08530661,0.0002258798],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5154543,0.002782325,0.3179919,0.001688742,0.0007478423,0.001542318,0.05122355,0.09587017,0.01269877],"genre_scores_gemma":[0.4216682,0.001233013,0.3925918,0.0003998814,0.00008082755,0.001821498,0.1686393,0.003038796,0.01052666],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01544665,"threshold_uncertainty_score":0.0307135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01645178177001421,"score_gpt":0.2612231025323662,"score_spread":0.244771320762352,"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."}}