{"id":"W2795970199","doi":"10.1002/pro.3406","title":"ProtaBank: A repository for protein design and engineering data","year":2018,"lang":"en","type":"article","venue":"Protein Science","topic":"Protein Structure and Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":71,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"National Institute of General Medical Sciences; National Institutes of Health","keywords":"Computer science; Suite; Relational database; Data mining; Visualization; Data type; Data visualization; Database; Information retrieval; Programming language","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":[],"consensus_categories":[],"category_scores_codex":[0.005843921,0.003165571,0.003728502,0.008171886,0.001956683,0.009419942,0.01008018,0.002410079,0.04620957],"category_scores_gemma":[0.01863101,0.00238037,0.002549927,0.01456865,0.001114645,0.009277388,0.007887968,0.004942061,0.07182841],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002219711,"about_ca_system_score_gemma":0.008190338,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002821427,"about_ca_topic_score_gemma":0.004206476,"domain_scores_codex":[0.9950506,0.0005650911,0.001039133,0.0009149816,0.002075758,0.0003544362],"domain_scores_gemma":[0.9869015,0.002238881,0.001944979,0.005414693,0.002301912,0.001198163],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001426835,0.000268104,0.003142075,0.008530768,0.0004715956,0.0007046513,0.0003807438,0.004678544,0.01152499,0.02030187,0.8207219,0.1278479],"study_design_scores_gemma":[0.00028101,0.0001168997,0.001579706,0.0006022431,0.0002002064,0.0007284619,0.0001081741,0.007551203,0.01334087,0.02142953,0.9538327,0.0002289448],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"software","genre_scores_codex":[0.005115442,0.005728337,0.174886,0.001158271,0.0008828425,0.0005211571,0.5409341,0.2465331,0.02424073],"genre_scores_gemma":[0.01007916,0.003127264,0.1071586,0.0005396061,0.0001186921,0.0006836533,0.8515977,0.01924041,0.007454865],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.04620957,"threshold_uncertainty_score":0.1545864,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01713230434062571,"score_gpt":0.2605485118772515,"score_spread":0.2434162075366258,"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."}}