{"id":"W4416571000","doi":"10.1038/s41592-025-02939-1","title":"Helixer: ab initio prediction of primary eukaryotic gene models combining deep learning and a hidden Markov model","year":2025,"lang":"en","type":"article","venue":"Nature Methods","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"Bureau de Coopération Interuniversitaire","funders":"German Network for Bioinformatics Infrastructure; Heinrich-Heine-Universität Düsseldorf; Forschungszentrum Jülich; Bundesministerium für Bildung und Forschung; Deutsche Forschungsgemeinschaft","keywords":"Hidden Markov model; Gene prediction; Identification (biology); Deep learning; Genome; Annotation; Software; Gene Annotation; Genome project","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005204708,0.0001513338,0.0002485071,0.0000771953,0.0001026919,0.00001193322,0.0001065523,0.0003372871,0.000001276805],"category_scores_gemma":[0.0001198106,0.0001491981,0.00006993535,0.0001047267,0.00007178212,0.000001592707,0.0002300466,0.0002948757,8.838932e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001545071,"about_ca_system_score_gemma":0.00006334971,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003629229,"about_ca_topic_score_gemma":0.00000128628,"domain_scores_codex":[0.9990247,0.000180851,0.0002269448,0.0003185825,0.00008370612,0.0001651879],"domain_scores_gemma":[0.9995087,0.00006392774,0.00008682457,0.0001838525,0.0001188997,0.00003772723],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001115775,0.00003713641,0.004069281,0.0001200461,0.0002264035,7.492221e-7,0.00018303,0.002891974,0.9074408,0.000416518,0.0001332893,0.08436923],"study_design_scores_gemma":[0.003622962,0.0007458536,0.03736551,0.0001789105,0.0005794152,0.00003927448,0.0004936488,0.3923874,0.5279452,0.03204556,0.003795538,0.0008006693],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6453352,0.0512544,0.2921391,0.0000923778,0.0003053831,0.0002476495,0.00003029223,0.00001238823,0.01058322],"genre_scores_gemma":[0.6854388,0.00264818,0.3112735,0.0002128492,0.0000475619,0.00001521891,0.00004149739,0.00001393011,0.0003084972],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3894955,"threshold_uncertainty_score":0.6084123,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01426356486179845,"score_gpt":0.2989379692566936,"score_spread":0.2846744043948951,"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."}}