{"id":"W4411622822","doi":"10.1093/gigascience/giaf070","title":"Extraction of biological terms using large language models enhances the usability of metadata in the BioSample database","year":2025,"lang":"en","type":"article","venue":"GigaScience","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute of Genetics; National Bioscience Database Center; Japan Society for the Promotion of Science; Japan Science and Technology Corporation","keywords":"Metadata; Computer science; Data curation; Information retrieval; World Wide Web; Usability; Data science; Database","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":[0.005180649,0.001870102,0.001291574,0.005719865,0.0009539151,0.003143613,0.001981616,0.001125957,0.002434405],"category_scores_gemma":[0.02039512,0.0005544748,0.002861508,0.002885567,0.0006935354,0.004349273,0.0036675,0.002230822,0.003205611],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001345521,"about_ca_system_score_gemma":0.004080608,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007058961,"about_ca_topic_score_gemma":0.01327192,"domain_scores_codex":[0.994651,0.001626133,0.0009918237,0.001258466,0.001254106,0.000218378],"domain_scores_gemma":[0.9852825,0.008973144,0.001028949,0.002495877,0.001972153,0.0002473888],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001031846,0.0007877296,0.02973163,0.00644102,0.0006110443,0.002190717,0.003764562,0.02271346,0.1548468,0.02043832,0.07248763,0.6849553],"study_design_scores_gemma":[0.0002899482,0.0004129044,0.021045,0.0008508182,0.0007742526,0.001711123,0.003077793,0.5167012,0.1561637,0.04409119,0.25446,0.0004221213],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09510799,0.002652165,0.7900223,0.002352632,0.0003157212,0.001221128,0.04822694,0.05423869,0.005862403],"genre_scores_gemma":[0.1219037,0.0008925275,0.7748445,0.0009737685,0.0001136915,0.001113503,0.09583174,0.002228538,0.002097985],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007058961,"threshold_uncertainty_score":0.02739817,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0670437112549849,"score_gpt":0.3748008608905625,"score_spread":0.3077571496355777,"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."}}