{"id":"W3139830754","doi":"10.1101/2021.04.05.438442","title":"The accuracy of NMR protein structures in the Protein Data Bank","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Protein Structure and Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"College of Family Physicians of Canada; University of Toronto","funders":"Core Research for Evolutional Science and Technology; Japan Science and Technology Agency; Biotechnology and Biological Sciences Research Council","keywords":"Protein Data Bank; Ramachandran plot; Protein Data Bank (RCSB PDB); Residue (chemistry); Chemistry; Data bank; Chemical shift; Protein structure; Biological system; Crystallography; Computer science; Stereochemistry; Biology; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01474932,0.0008452809,0.001465438,0.004595508,0.000849104,0.003099501,0.001269554,0.001016996,0.001905095],"category_scores_gemma":[0.04608213,0.0005494716,0.0006841485,0.004471208,0.0009132401,0.00214739,0.001629223,0.001199412,0.002452012],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001260964,"about_ca_system_score_gemma":0.001048508,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004390812,"about_ca_topic_score_gemma":0.002974976,"domain_scores_codex":[0.983569,0.0049461,0.002125831,0.002833426,0.005818107,0.0007074837],"domain_scores_gemma":[0.9751091,0.009092452,0.002308323,0.008859686,0.004227526,0.0004029491],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.005139839,0.0004320978,0.2632026,0.001228941,0.002089298,0.001031043,0.00163628,0.2739613,0.07892467,0.03747479,0.0471435,0.2877358],"study_design_scores_gemma":[0.0001731944,0.0002591675,0.08125426,0.0002889296,0.0003028002,0.0009748221,0.000424049,0.7423936,0.0982862,0.02644773,0.0489529,0.0002423677],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7431584,0.00528559,0.18394,0.001080449,0.0004762201,0.0001264609,0.04046533,0.01239504,0.01307251],"genre_scores_gemma":[0.8391991,0.0009557013,0.1077139,0.0001886954,0.00007231608,0.0001143011,0.04891689,0.001191664,0.001647315],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9852507,"threshold_uncertainty_score":0.07800281,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01421647741981422,"score_gpt":0.243449594578087,"score_spread":0.2292331171582727,"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."}}