{"id":"W4404404643","doi":"10.1101/2024.11.13.621715","title":"Accurate structure prediction of immune proteins using parameter-efficient transfer learning","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"vaccines and immunoinformatics approaches","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Transfer of learning; Computer science; Immune system; Artificial intelligence; Computational biology; Biology; Immunology","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":[0.0003533746,0.000489054,0.0004682807,0.000171623,0.0001140961,0.0001398418,0.0003724371,0.0006639635,0.00001377572],"category_scores_gemma":[0.0001109103,0.0004688195,0.0002761478,0.0002555401,0.00007494369,0.000007858075,0.0005735375,0.0008171981,0.000003670503],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006756802,"about_ca_system_score_gemma":0.0003918957,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003335634,"about_ca_topic_score_gemma":5.514941e-7,"domain_scores_codex":[0.9978665,0.00008569773,0.0007884087,0.000620072,0.0002540865,0.0003852677],"domain_scores_gemma":[0.9984869,0.000009317152,0.0002839708,0.0008356868,0.0002912516,0.00009292076],"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":[0.0000470585,0.00005741138,0.0002597619,0.00103004,0.0002968609,0.000002476706,0.00002754416,0.02129466,0.9769055,0.00006596845,0.000009888587,0.000002873418],"study_design_scores_gemma":[0.00037331,0.0001675093,0.004110794,0.0004624565,0.0002071773,9.789952e-8,0.00001496056,0.04707381,0.946587,0.000001892984,0.0005495663,0.0004514247],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.98746,0.003095259,0.007343156,0.00002060162,0.0006722331,0.0008298085,0.0004971906,0.0000772985,0.000004514256],"genre_scores_gemma":[0.995876,0.0001782682,0.003485943,0.00001279676,0.0002601294,0.00006169012,0.000006471632,0.000113288,0.000005396447],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03031846,"threshold_uncertainty_score":0.9997764,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01482646452006505,"score_gpt":0.2159739988016856,"score_spread":0.2011475342816205,"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."}}