{"id":"W3126773939","doi":"10.1038/s41467-021-21194-4","title":"RNA secondary structure prediction using deep learning with thermodynamic integration","year":2021,"lang":"en","type":"article","venue":"Nature Communications","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":496,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute of Genetics; Japan Society for the Promotion of Science; Ministry of Education, Culture, Sports, Science and Technology; Research Organization of Information and Systems","keywords":"Overfitting; Nucleic acid secondary structure; Computer science; Parameterized complexity; RNA; Artificial intelligence; Protein secondary structure; Robustness (evolution); Regularization (linguistics); Non-coding RNA; Machine learning; Deep learning; Folding (DSP implementation); Nucleic acid structure; Artificial neural network; Computational biology; Algorithm; Biology; Gene; Genetics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007379917,0.0008517023,0.0006738012,0.0005674182,0.0003344542,0.0007091117,0.000850385,0.0007388798,0.001053742],"category_scores_gemma":[0.001501144,0.0004953534,0.0006401656,0.000400389,0.0004582498,0.001023313,0.0009741789,0.00131173,0.0003315342],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00107135,"about_ca_system_score_gemma":0.001259429,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004465153,"about_ca_topic_score_gemma":0.005704025,"domain_scores_codex":[0.9997976,0.00004821536,0.00001144921,0.00006224109,0.0000562303,0.00002420922],"domain_scores_gemma":[0.9995372,0.0002065154,0.00005956471,0.00005777131,0.00009274414,0.00004617607],"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":[0.00006553326,0.0000882533,0.001092658,0.00002885365,0.0000392601,0.0000277886,0.00001498668,0.9334636,0.007770083,0.002499745,0.0004915299,0.05441766],"study_design_scores_gemma":[0.000001590728,0.00000406799,0.0000276517,6.070152e-7,7.63225e-7,0.000001425549,5.340863e-7,0.9985321,0.000777884,0.000616867,0.00003548754,0.000001085101],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1443747,0.000346155,0.84948,0.0003123281,0.00005442948,0.00005358557,0.0001582499,0.0030163,0.002204245],"genre_scores_gemma":[0.7799839,0.0001292654,0.2170155,0.0001718356,0.00003554595,0.00009612702,0.000537866,0.0001440951,0.001885831],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004465153,"threshold_uncertainty_score":0.00887835,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008972905914192607,"score_gpt":0.2518728836326812,"score_spread":0.2428999777184886,"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."}}