{"id":"W4414598938","doi":"10.1101/2025.09.10.25334730","title":"EAGLE-AI: A large language model workflow for automated extraction and scoring of literature evidence linking genes to autism spectrum disorder","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; SickKids Foundation; Hospital for Sick Children","funders":"University of Toronto; Hospital for Sick Children; Autism Speaks","keywords":"Workflow; Autism; Autism spectrum disorder; Set (abstract data type); Data curation; Component (thermodynamics); Data set; Genomics","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":[0.00980639,0.001894788,0.001148279,0.009329503,0.001154435,0.003944427,0.002098085,0.001125641,0.01763907],"category_scores_gemma":[0.02194284,0.0009215701,0.00248098,0.003520336,0.0004968141,0.00238584,0.003408811,0.001432583,0.01153262],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001274734,"about_ca_system_score_gemma":0.005977998,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00525235,"about_ca_topic_score_gemma":0.01412712,"domain_scores_codex":[0.9964839,0.0009542431,0.0006577226,0.0008002055,0.0009576417,0.0001461852],"domain_scores_gemma":[0.9874517,0.006919471,0.001011398,0.001850249,0.002205634,0.0005614851],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001328008,0.000269501,0.009967417,0.009042396,0.001534551,0.001672104,0.001695765,0.01171712,0.06623996,0.0220419,0.476614,0.3978772],"study_design_scores_gemma":[0.0006362287,0.0002912612,0.01008985,0.001233158,0.0006875891,0.00184361,0.000981123,0.1719091,0.1024211,0.06327002,0.6461601,0.0004770475],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.009843997,0.001803602,0.5914395,0.00225586,0.0003168551,0.001107555,0.1292935,0.2573861,0.00655299],"genre_scores_gemma":[0.03992851,0.0007984145,0.827262,0.0007017289,0.0001246032,0.0009008757,0.1172874,0.008208789,0.004787701],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01763907,"threshold_uncertainty_score":0.0590086,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02099555704033951,"score_gpt":0.3390197854823722,"score_spread":0.3180242284420327,"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."}}