{"id":"W4412105042","doi":"10.1101/2025.07.05.662870","title":"mRNABench: A curated benchmark for mature mRNA property and function prediction","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; Vector Institute; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; National Cancer Institute; National Science Foundation; National Institutes of Health; Government of Canada; Canadian Institute for Advanced Research; Memorial Sloan-Kettering Cancer Center","keywords":"Benchmark (surveying); Property (philosophy); Function (biology); Messenger RNA; Computational biology; Computer science; Biology; Evolutionary biology; Geography; Cartography; Genetics; Gene; Philosophy","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":[0.003547931,0.003598033,0.001262483,0.002066549,0.001054034,0.001609889,0.003628795,0.00254492,0.004717411],"category_scores_gemma":[0.009624062,0.0007812916,0.00175156,0.00304826,0.0008916291,0.001754706,0.001428859,0.002072379,0.00636459],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001486166,"about_ca_system_score_gemma":0.002089089,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008015883,"about_ca_topic_score_gemma":0.01347001,"domain_scores_codex":[0.9977559,0.0005750692,0.0002117887,0.000699819,0.000586262,0.0001710308],"domain_scores_gemma":[0.9965681,0.001746113,0.0001889834,0.0007344823,0.00061092,0.0001514401],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001521609,0.0008645083,0.01309951,0.006462173,0.001046766,0.000563138,0.000266525,0.2356699,0.02520412,0.007190491,0.5409558,0.1671554],"study_design_scores_gemma":[0.0004829195,0.0007747881,0.007491371,0.0002897567,0.0001835322,0.0005235387,0.0002070589,0.824949,0.03944388,0.0131708,0.1123545,0.0001289988],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.2834618,0.01881316,0.2008843,0.00268505,0.001633078,0.001110428,0.2926132,0.164656,0.03414303],"genre_scores_gemma":[0.1501843,0.002414264,0.1596103,0.0008646533,0.0001128958,0.001252217,0.6711217,0.007883303,0.006556457],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.008015883,"threshold_uncertainty_score":0.01876348,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0105795069096468,"score_gpt":0.209182925187773,"score_spread":0.1986034182781262,"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."}}