{"id":"W4416205407","doi":"10.1093/bib/bbaf590","title":"M3Site: multiclass multimodal learning for protein active site identification and classification","year":2025,"lang":"en","type":"article","venue":"Briefings in Bioinformatics","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; Cegep Edouard Montpetit","funders":"Fundamental Research Funds for the Central Universities; Wuhan University; Renmin Hospital of Wuhan University; National Natural Science Foundation of China; Innovative Research Group Project of the National Natural Science Foundation of China","keywords":"UniProt; Feature (linguistics); Source code; Multiclass classification; Active learning (machine learning); Graph; Identification (biology); Limiting; Binary classification","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":[],"consensus_categories":[],"category_scores_codex":[0.0006967707,0.000128245,0.0001507185,0.000297135,0.0001678621,0.0002973389,0.0002840823,0.00008094883,5.496547e-7],"category_scores_gemma":[0.000813849,0.0001390885,0.00003902699,0.0005184875,0.00005622833,0.001161172,0.0001810862,0.0001657373,0.00000810218],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000138496,"about_ca_system_score_gemma":0.0001050794,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003243343,"about_ca_topic_score_gemma":0.000009984019,"domain_scores_codex":[0.9988105,0.00007322351,0.0004915562,0.0002423374,0.0001814213,0.0002009439],"domain_scores_gemma":[0.9989206,0.0004135844,0.0002443758,0.0002252949,0.0001603529,0.00003580246],"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":[0.00007116366,0.000123161,0.001296875,0.0005260676,0.00003288883,7.769422e-7,0.008574129,0.02544317,0.01080313,0.209681,0.0002838974,0.7431638],"study_design_scores_gemma":[0.0005828796,0.00001945772,0.0485406,0.00008189958,0.000004871908,0.000001737657,0.0001536939,0.9394814,0.002976493,0.006247818,0.00177599,0.0001331558],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1130064,0.00001948832,0.8839266,0.001706452,0.00008462981,0.0007227883,0.000004841906,0.00008590227,0.0004429094],"genre_scores_gemma":[0.5425429,0.0000064845,0.4566584,0.0003258267,0.00001160256,0.0001464918,0.00003115552,0.000006530044,0.0002705876],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9140382,"threshold_uncertainty_score":0.5671866,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01819747715008626,"score_gpt":0.3033901452785213,"score_spread":0.285192668128435,"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."}}