{"id":"W4405778687","doi":"10.1101/2024.12.15.628585","title":"EnzymeCAGE: A Geometric Foundation Model for Enzyme Retrieval with Evolutionary Insights","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Microbial Metabolic Engineering and Bioproduction","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Foundation (evidence); Computer science; Artificial intelligence; Geography; Archaeology","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":[0.0003216176,0.000502272,0.0003614675,0.0004223747,0.0001317149,0.0001390105,0.0003053451,0.0006288542,0.000004872156],"category_scores_gemma":[0.0002129272,0.0004765426,0.0001705812,0.0006345563,0.00007886027,0.00001076392,0.000345906,0.0004675376,0.0000216444],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001388957,"about_ca_system_score_gemma":0.0006858502,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009314071,"about_ca_topic_score_gemma":0.000001584883,"domain_scores_codex":[0.9977121,0.00003957051,0.0003793437,0.001206633,0.0002553028,0.0004070729],"domain_scores_gemma":[0.9981731,0.00000753947,0.0001965572,0.0009068698,0.0005693629,0.0001465242],"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":[0.0001702515,0.00007318737,0.00001689945,0.0005431178,0.0002066216,0.000003541129,0.000004770848,0.006852491,0.9901054,0.0003490747,0.001671075,0.000003572769],"study_design_scores_gemma":[0.0005440874,0.000200365,0.000586041,0.0002217058,0.000268261,1.464421e-7,0.000001129718,0.02172176,0.9522362,0.00001361676,0.02334189,0.0008647786],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8719323,0.009368504,0.1149999,0.0001641196,0.001809685,0.001158195,0.0002987918,0.0002626685,0.000005912836],"genre_scores_gemma":[0.9823728,0.0004783234,0.01471048,0.00006570343,0.001901485,0.0001807174,0.00001381313,0.0001562396,0.0001204051],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1104406,"threshold_uncertainty_score":0.9997686,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009832383824401011,"score_gpt":0.2099745982871137,"score_spread":0.2001422144627126,"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."}}