{"id":"W4286797666","doi":"10.1101/2022.07.23.501263","title":"Predicting the specific substrate for transmembrane transport proteins using BERT language model","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"UniProt; Computer science; Classifier (UML); Transmembrane protein; Artificial intelligence; Language model; Machine learning; Chemistry; Biochemistry","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001015074,0.0005832515,0.0004346977,0.0001073974,0.0004739053,0.0001145057,0.0009479729,0.0004839998,0.00003799145],"category_scores_gemma":[0.00008056808,0.0005402139,0.0003425823,0.0002050678,0.000133287,0.00001319415,0.000267782,0.0009849907,0.000002195718],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001030495,"about_ca_system_score_gemma":0.000613042,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004911857,"about_ca_topic_score_gemma":0.000008496289,"domain_scores_codex":[0.9973497,0.0001151711,0.0006875445,0.0008189878,0.0004064437,0.0006221958],"domain_scores_gemma":[0.9977136,0.00002930731,0.0004788855,0.0014354,0.0001941487,0.0001486249],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001256943,0.00008340768,0.002977948,0.0007344951,0.0001954911,0.000008911119,0.0001343028,0.06650686,0.9290255,0.000133293,0.00007224621,0.00000185362],"study_design_scores_gemma":[0.001725472,0.0002776378,0.008337133,0.0002879398,0.0004113045,2.812422e-7,0.00008531044,0.2943009,0.6782281,0.000003308688,0.01441235,0.001930272],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9397758,0.001104436,0.05526724,0.0001339505,0.0004734979,0.002102999,0.0009613631,0.0001454156,0.00003532209],"genre_scores_gemma":[0.9705583,0.0001781485,0.02767267,0.0001826308,0.0006028299,0.0005565436,0.00001700043,0.0002014646,0.00003034875],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2507974,"threshold_uncertainty_score":0.999705,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0152082455531926,"score_gpt":0.2387239810005427,"score_spread":0.2235157354473502,"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."}}