{"id":"W4225093366","doi":"10.1101/2022.01.27.478113","title":"Integrated pretraining with evolutionary information to improve RNA secondary structure prediction","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Institut de Valorisation des Données; Compute Canada","keywords":"Computer science; Scalability; Nucleic acid secondary structure; RNA; Source code; Protein secondary structure; Set (abstract data type); Folding (DSP implementation); Evolutionary algorithm; Machine learning; Task (project management); Software; Artificial intelligence; Theoretical computer science; Computational biology; Biology; Database; Engineering; Genetics","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":[0.001064485,0.001532353,0.0008879119,0.0007053047,0.0003677482,0.0006457554,0.001598437,0.001397523,0.003187601],"category_scores_gemma":[0.003508421,0.000555015,0.0006257888,0.0005533902,0.0004853357,0.001211156,0.001022832,0.002454373,0.001606892],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000574538,"about_ca_system_score_gemma":0.001047914,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004363362,"about_ca_topic_score_gemma":0.008333582,"domain_scores_codex":[0.9995435,0.00009674343,0.00002529132,0.0001490637,0.0001126555,0.00007277542],"domain_scores_gemma":[0.9985994,0.0007325159,0.00007913511,0.000204415,0.0003150593,0.00006940446],"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":[0.00023212,0.0003615734,0.002909645,0.0001554793,0.0001080078,0.0001456493,0.00009001612,0.605377,0.02200134,0.002366357,0.008656288,0.3575965],"study_design_scores_gemma":[0.00000935348,0.00004236905,0.0002372244,0.000008434199,0.000008834481,0.00001361041,0.00000578634,0.9944522,0.003665011,0.0009784381,0.0005739655,0.000004766446],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1632648,0.001719736,0.8118629,0.0009560276,0.0003890053,0.0001412251,0.0005838643,0.01379779,0.007284532],"genre_scores_gemma":[0.6273342,0.0005657941,0.358893,0.0009825404,0.0002357161,0.0002641341,0.003477847,0.0006729385,0.007573851],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004363362,"threshold_uncertainty_score":0.01066357,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00529099688720107,"score_gpt":0.1877260198177303,"score_spread":0.1824350229305292,"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."}}