{"id":"W4312094950","doi":"","title":"Enzyme Activity Prediction of Sequence Variants on Novel Substrates using Improved Substrate Encodings and Convolutional Pooling.","year":2022,"lang":"en","type":"article","venue":"PubMed","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Pooling; Computer science; Artificial intelligence; Workflow; Machine learning; Computational biology; Data mining; Biology; Database","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":[0.0003979247,0.001149991,0.0005493365,0.0009986588,0.0001107754,0.000482181,0.0006927476,0.0005175365,0.0007295145],"category_scores_gemma":[0.0007734148,0.0002639757,0.0007463202,0.0008394083,0.0001732945,0.0008148017,0.0005469418,0.0007054022,0.0005578452],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005404037,"about_ca_system_score_gemma":0.0003749763,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004649501,"about_ca_topic_score_gemma":0.008038372,"domain_scores_codex":[0.9998197,0.00002645171,0.00001175181,0.00007776517,0.00003928388,0.00002497635],"domain_scores_gemma":[0.9997764,0.00005855255,0.00005215978,0.00004285283,0.00004784378,0.00002215552],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001647603,0.0007005738,0.05311115,0.0004957126,0.0006523269,0.0005210553,0.0001343161,0.2253538,0.1186688,0.002935954,0.01324767,0.582531],"study_design_scores_gemma":[0.00001677285,0.00009435246,0.0037169,0.000006780982,0.00004201509,0.0000907339,0.00001489652,0.9803633,0.01345598,0.001210553,0.0009755716,0.00001212687],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6613484,0.003639711,0.3142479,0.0005359645,0.00009967841,0.00009631512,0.006231561,0.01048157,0.003318869],"genre_scores_gemma":[0.8987851,0.0006450369,0.08659518,0.0001119976,0.00003080294,0.00004708755,0.01151045,0.0001097603,0.002164568],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004649501,"threshold_uncertainty_score":0.009244859,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08933320113248036,"score_gpt":0.2743013484337087,"score_spread":0.1849681473012283,"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."}}