{"id":"W4386305688","doi":"10.1093/bib/bbad319","title":"Sequence-based prediction model of protein crystallization propensity using machine learning and two-level feature selection","year":2023,"lang":"en","type":"article","venue":"Briefings in Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":63,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Science and Technology Council","keywords":"Feature selection; Pipeline (software); Protein crystallization; Computer science; Support vector machine; Hyperparameter; Dimensionality reduction; Artificial intelligence; Machine learning; Sequence (biology); Selection (genetic algorithm); Linear discriminant analysis; Feature (linguistics); Data mining; Crystallization; Engineering; Chemistry","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005305664,0.0001832804,0.000191499,0.0002110079,0.0001403863,0.00004371387,0.00009849526,0.0001977061,0.000002101555],"category_scores_gemma":[0.0003643308,0.0001829305,0.00004035017,0.000434508,0.00008552252,0.00003873448,0.000116252,0.0002670627,0.000001356307],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004423683,"about_ca_system_score_gemma":0.0001400993,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001088223,"about_ca_topic_score_gemma":0.00006015053,"domain_scores_codex":[0.9988533,0.0000489573,0.0004570509,0.000157701,0.0002398286,0.0002431273],"domain_scores_gemma":[0.9992986,0.00001239082,0.0003530903,0.0001372947,0.0001492585,0.00004932182],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001001975,0.00002978482,0.03027627,0.0005892869,0.0000227701,5.921248e-7,0.000537308,0.5871272,0.3789211,0.0001374024,0.00007744306,0.002180617],"study_design_scores_gemma":[0.000698462,0.0001359181,0.0009500949,0.0001281858,0.00001255729,0.00001435346,0.00004460779,0.9645515,0.03285777,0.00004445423,0.0003898332,0.0001722783],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8651489,0.000026349,0.1337868,0.0001172099,0.00002658541,0.0005256039,0.0000650793,0.00007811371,0.0002253407],"genre_scores_gemma":[0.8175345,0.00005515005,0.1811568,0.0001591997,0.00003276382,0.00001743902,0.0007096349,0.00003255545,0.0003020229],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3774243,"threshold_uncertainty_score":0.7459691,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03691560672882826,"score_gpt":0.2716449149048197,"score_spread":0.2347293081759914,"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."}}