{"id":"W4365444089","doi":"10.1093/bioinformatics/btad189","title":"Structure-aware protein self-supervised learning","year":2023,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":48,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Artificial intelligence; Machine learning; Pairwise comparison; Leverage (statistics); Graph; Supervised learning; Artificial neural network; Theoretical computer science","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":[0.000291348,0.0002581594,0.0001873662,0.0001420459,0.0002313372,0.0000937322,0.000363216,0.0002443859,0.00006765454],"category_scores_gemma":[0.0002518866,0.0002354872,0.0001015516,0.0003807853,0.00004984011,0.00001907642,0.0002968168,0.0003010702,0.000483977],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002224505,"about_ca_system_score_gemma":0.0001001344,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004326531,"about_ca_topic_score_gemma":0.000004635111,"domain_scores_codex":[0.9985185,0.00005019768,0.0004794702,0.0001658122,0.0003198718,0.0004661755],"domain_scores_gemma":[0.9990985,0.00001613426,0.0001959459,0.0004583086,0.000101936,0.0001291689],"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":[0.0005982181,0.0004598483,0.07369094,0.009458076,0.001984182,0.00007358062,0.0310849,0.06977375,0.4023435,0.005234156,0.1960979,0.2092009],"study_design_scores_gemma":[0.001439024,0.0006471485,0.002101151,0.00007915552,0.00003981037,0.0000572198,0.001687671,0.7189814,0.03385085,0.00011584,0.2400074,0.0009933145],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9817455,0.00003862472,0.006539315,0.0002059358,0.0002849175,0.0007280287,0.00005375586,0.0007627664,0.009641139],"genre_scores_gemma":[0.9406601,0.00007629182,0.05331495,0.0004199187,0.0003403173,0.00004087335,0.001653968,0.00007954453,0.00341408],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6492077,"threshold_uncertainty_score":0.960289,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007098345358430243,"score_gpt":0.2367498723285093,"score_spread":0.2296515269700791,"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."}}