{"id":"W1571001643","doi":"10.1007/11424918_17","title":"Fast Protein Superfamily Classification Using Principal Component Null Space Analysis","year":2005,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Windsor","funders":"","keywords":"Computer science; Principal component analysis; Artificial intelligence; Curse of dimensionality; Classifier (UML); Artificial neural network; Machine learning; Pattern recognition (psychology); Support vector machine; Data mining","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":[0.00138323,0.001592779,0.001725127,0.003885388,0.001091995,0.002086631,0.002315908,0.001207239,0.01067835],"category_scores_gemma":[0.00352057,0.0007136657,0.001647068,0.003165171,0.00045752,0.002067799,0.002302242,0.001809163,0.01104762],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005208022,"about_ca_system_score_gemma":0.001239428,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002277684,"about_ca_topic_score_gemma":0.002763142,"domain_scores_codex":[0.9990343,0.0001742144,0.00006221616,0.0001911563,0.0004513699,0.00008683757],"domain_scores_gemma":[0.9980497,0.0006305001,0.00009375619,0.000456258,0.0006765082,0.00009328688],"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":[0.0002683489,0.00009094135,0.0007250687,0.0001778446,0.00008350534,0.00005517913,0.00006955288,0.007240638,0.01422855,0.00601918,0.01891049,0.9521307],"study_design_scores_gemma":[0.0001005114,0.0001036612,0.002681251,0.00005292045,0.00008144523,0.0003473566,0.0001289886,0.9038341,0.02671131,0.04703491,0.01884757,0.00007606836],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008290672,0.0005614288,0.9737957,0.0001302516,0.0001222344,0.0001074853,0.0009615573,0.01472644,0.001304299],"genre_scores_gemma":[0.05196559,0.0004008663,0.9379877,0.00006841971,0.0000657977,0.0002568806,0.004320287,0.0008564731,0.004077995],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01067835,"threshold_uncertainty_score":0.03572261,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01715153564647308,"score_gpt":0.2630029553141384,"score_spread":0.2458514196676653,"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."}}