{"id":"W4390023653","doi":"10.1101/2023.12.19.572475","title":"Inverse folding of protein complexes with a structure-informed language model enables unsupervised antibody evolution","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institutes of Health Research; Virginia and D.K. Ludwig Fund for Cancer Research","keywords":"Computer science; Computational biology; Artificial intelligence; Protein folding; Inverse; Sequence (biology); Protein design; Folding (DSP implementation); Protein engineering; Protein sequencing; Language model; Protein structure; Function (biology); Peptide sequence; Machine learning; Biology; Mathematics; Genetics; Biochemistry; Engineering; Gene","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003051403,0.0005148444,0.0005248663,0.0002194218,0.0001195736,0.00007303784,0.0005411671,0.0005883352,0.00001353481],"category_scores_gemma":[0.0001682449,0.0004818798,0.0001510868,0.0002661186,0.0001287508,0.00001645005,0.0004490497,0.0002636749,0.000006055681],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000906174,"about_ca_system_score_gemma":0.0007732928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001483268,"about_ca_topic_score_gemma":0.00004243082,"domain_scores_codex":[0.9978169,0.0000852632,0.0004626309,0.0007914571,0.0003660285,0.000477741],"domain_scores_gemma":[0.9979154,0.00001625205,0.0004475497,0.00114524,0.0003214074,0.0001541868],"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":[0.0001463921,0.00003956114,0.0007162938,0.0007426159,0.0001985651,0.00001177568,0.00002188047,0.00178031,0.9958355,0.0004712643,0.0000345626,0.000001267184],"study_design_scores_gemma":[0.0005704829,0.0001415159,0.001751774,0.0006186433,0.00008390879,2.932622e-8,0.00002588404,0.002187096,0.9939123,0.00002734817,0.00008780247,0.000593197],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.976993,0.0005124148,0.02044188,0.00003140592,0.0001432288,0.001158985,0.0005784573,0.0001345617,0.000006088133],"genre_scores_gemma":[0.9771842,0.00008176186,0.0221694,0.00003412186,0.0001778849,0.0001695964,0.000009906232,0.000143602,0.0000294677],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.001923193,"threshold_uncertainty_score":0.9997633,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01502527081235531,"score_gpt":0.2380155161854131,"score_spread":0.2229902453730578,"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."}}