{"id":"W4395082224","doi":"10.1002/prot.26694","title":"Exploiting protein language models for the precise classification of ion channels and ion transporters","year":2024,"lang":"en","type":"article","venue":"Proteins Structure Function and Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Robustness (evolution); Computer science; Random forest; Generalization; Artificial intelligence; Machine learning; Biology; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002278957,0.000147379,0.0001120988,0.00006618461,0.0001269207,0.00008315987,0.00007310859,0.0001385628,0.000005479808],"category_scores_gemma":[0.00004850917,0.00009739849,0.00004973435,0.00008712497,0.00007636856,0.00004072393,0.00002991002,0.0001254002,3.455623e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007555555,"about_ca_system_score_gemma":0.00002887579,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000598044,"about_ca_topic_score_gemma":0.000007041843,"domain_scores_codex":[0.9992589,0.00001574069,0.0003292086,0.0001310001,0.0001301073,0.0001350883],"domain_scores_gemma":[0.9995674,0.00002111924,0.0001402976,0.0001710627,0.0000601266,0.00003998749],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004553685,0.00002459556,0.0001415227,0.004958449,0.0002117699,3.072369e-7,0.007734193,0.002328004,0.6922941,0.007971707,0.000335576,0.2835444],"study_design_scores_gemma":[0.0007043277,0.0006482533,0.0005079102,0.0002523428,0.0001035173,0.00003082087,0.002939227,0.9012288,0.08632658,0.001571645,0.005392452,0.0002940665],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4086799,0.0009330243,0.5878195,0.0002946854,0.0003158481,0.00162605,0.00009392826,0.00004613598,0.0001909107],"genre_scores_gemma":[0.9870985,0.0000837725,0.01213149,0.00005297028,0.0002004479,0.00009085734,0.0002129445,0.00001588436,0.0001131276],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8989009,"threshold_uncertainty_score":0.3971795,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01184212033902125,"score_gpt":0.2373179589850951,"score_spread":0.2254758386460739,"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."}}