{"id":"W3088578860","doi":"10.1016/j.cels.2020.08.016","title":"Fast and Flexible Protein Design Using Deep Graph Neural Networks","year":2020,"lang":"en","type":"article","venue":"Cell Systems","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":230,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Fonds de recherche du Québec – Nature et technologies; Canadian Institutes of Health Research; Nvidia","keywords":"Computer science; In silico; Benchmark (surveying); Protein design; Graph; Artificial neural network; Constraint (computer-aided design); Algorithm; Protein structure prediction; Protein sequencing; Sequence (biology); Protein structure; Theoretical computer science; Artificial intelligence; Peptide sequence; Mathematics; Biology; Gene; Genetics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005454061,0.001192401,0.0008363337,0.0006189361,0.0004623666,0.0007407318,0.001187515,0.001165843,0.00400602],"category_scores_gemma":[0.001318302,0.0006637571,0.0009804631,0.000764244,0.0006440884,0.0009587822,0.0008988193,0.001343461,0.001125712],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001497812,"about_ca_system_score_gemma":0.001243767,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0047717,"about_ca_topic_score_gemma":0.01195139,"domain_scores_codex":[0.9996774,0.00007381273,0.00001388295,0.00009140246,0.0001019399,0.00004169071],"domain_scores_gemma":[0.9995373,0.0001992114,0.00005343039,0.00009170957,0.0000772174,0.00004120355],"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":[0.0001374914,0.0001119664,0.001133613,0.0002899445,0.00008389594,0.00013387,0.00005446695,0.8024563,0.01659422,0.02030407,0.008439207,0.1502609],"study_design_scores_gemma":[0.00002054463,0.00002763067,0.00006619366,0.000007500949,0.000005699829,0.00001393056,0.00000784422,0.9871193,0.001912711,0.009186118,0.001627815,0.000004694113],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09240169,0.001064184,0.8883299,0.0005920953,0.000127318,0.0001480053,0.0007558627,0.008006416,0.008574368],"genre_scores_gemma":[0.4268332,0.0005721434,0.5624157,0.0004446393,0.00003997991,0.000298473,0.00217558,0.0009570174,0.006263191],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0047717,"threshold_uncertainty_score":0.01340151,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01847110912025982,"score_gpt":0.2246390681826954,"score_spread":0.2061679590624356,"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."}}