{"id":"W4403908137","doi":"10.1038/s41598-024-77172-5","title":"StructmRNA a BERT based model with dual level and conditional masking for mRNA representation","year":2024,"lang":"en","type":"article","venue":"Scientific Reports","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Masking (illustration); Representation (politics); Computer science; Dual (grammatical number); Messenger RNA; Artificial intelligence; Computational biology; Biology; Gene; Genetics; Art","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":[0.0004764875,0.0009470314,0.0006340125,0.0004063621,0.0003753286,0.0008841922,0.001348975,0.001450194,0.004601069],"category_scores_gemma":[0.001390494,0.0004938189,0.0009421585,0.0003028033,0.0008112213,0.001386274,0.0008979402,0.001534085,0.001373574],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009913415,"about_ca_system_score_gemma":0.001071802,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004699092,"about_ca_topic_score_gemma":0.006019557,"domain_scores_codex":[0.9997595,0.00005538155,0.000009600092,0.00008851938,0.00004745001,0.00003947032],"domain_scores_gemma":[0.9995727,0.0002116329,0.0000504967,0.00004989774,0.00006928433,0.00004606678],"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":[0.0002819499,0.00006008312,0.001851916,0.00008958542,0.00004820393,0.0002098345,0.00009058214,0.8861514,0.01144719,0.05405648,0.003428429,0.04228432],"study_design_scores_gemma":[0.000005152617,0.00002704814,0.00007424499,0.000003194704,0.000006675553,0.00002479555,0.000003610093,0.9902654,0.0007741924,0.008009988,0.0007993318,0.00000633845],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04690384,0.0005228035,0.9432689,0.000908767,0.000181221,0.00005248689,0.0008022308,0.001845131,0.005514627],"genre_scores_gemma":[0.8143943,0.0005714516,0.1609344,0.0006501784,0.0001625651,0.0002974921,0.002135005,0.0004448543,0.0204098],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004699092,"threshold_uncertainty_score":0.01539207,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03179457955419966,"score_gpt":0.2786868528159258,"score_spread":0.2468922732617261,"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."}}