{"id":"W4416295244","doi":"10.1101/2025.11.13.688294","title":"TooTranslator: Zero-Shot Classification of Specific Substrates for Transport Proteins by Language Embedding Alignment of Proteins and Chemicals","year":2025,"lang":"","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada; Genome Canada","keywords":"Embedding; Transmembrane protein; Task (project management); Substrate (aquarium); Substrate specificity; Function (biology); Class (philosophy); Training set","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001183156,0.001246072,0.0009926964,0.0007413988,0.0003648745,0.001301422,0.001798585,0.001813504,0.002799296],"category_scores_gemma":[0.002667916,0.0004313758,0.00109879,0.0004748242,0.0005509342,0.002007752,0.00144684,0.001818122,0.001715402],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008839517,"about_ca_system_score_gemma":0.0009516212,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00341943,"about_ca_topic_score_gemma":0.003613515,"domain_scores_codex":[0.9994015,0.0002104915,0.00002371048,0.0001969608,0.00009379932,0.00007359599],"domain_scores_gemma":[0.9988006,0.0005686078,0.0001252897,0.0002193593,0.0001914363,0.0000946534],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001509203,0.0008771017,0.009514754,0.0004689947,0.0003965044,0.0005179837,0.0002021143,0.5621726,0.03820692,0.006594031,0.02557925,0.3539605],"study_design_scores_gemma":[0.00001004376,0.00006929944,0.0001721477,0.000005263536,0.000005104275,0.00002962119,0.00001293548,0.993417,0.004384685,0.001493753,0.0003934987,0.000006631567],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4179827,0.001697178,0.5430832,0.001342681,0.0003773765,0.0001604644,0.002536547,0.02916015,0.00365974],"genre_scores_gemma":[0.8326282,0.0002929727,0.1495683,0.0003942371,0.00009231456,0.000141618,0.008863491,0.0008723263,0.007146552],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00341943,"threshold_uncertainty_score":0.009364605,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01235022830420307,"score_gpt":0.2530394296372403,"score_spread":0.2406892013330372,"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."}}