{"id":"W4413112481","doi":"10.1038/s41597-025-05652-y","title":"Phased genome assemblies and pangenome graphs of human populations of Japan and Saudi Arabia","year":2025,"lang":"en","type":"article","venue":"Scientific Data","topic":"Genomics and Rare Diseases","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute of Genetics; Japan Science and Technology Agency; National Bioscience Database Center; King Salman Center for Disability Research; Dalian Institute of Chemical Physics; King Abdullah University of Science and Technology; Japan Society for the Promotion of Science; Ministry of Education, Culture, Sports, Science and Technology; King Fahad Medical City; Japan Agency for Medical Research and Development","keywords":"Biology; Haplotype; Reference genome; Genome; Population; Graph; Human genome; Computational biology; Gene; Genetics; Computer science; Allele; Theoretical computer science; Medicine","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000202322,0.00006585935,0.0001054681,0.0001092877,0.0001316802,0.00004269331,0.0002772025,0.00003535942,0.000007532715],"category_scores_gemma":[0.00004132495,0.00006214189,0.00002242438,0.0001447205,0.0003020445,0.000006219532,0.0003256365,0.00001825812,2.514404e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000161837,"about_ca_system_score_gemma":0.00004504315,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003131996,"about_ca_topic_score_gemma":0.0001057631,"domain_scores_codex":[0.9993139,0.00001694734,0.000165094,0.0003475788,0.00006300982,0.00009345474],"domain_scores_gemma":[0.9991056,0.000004687949,0.00007064833,0.0007173333,0.00006215886,0.00003962564],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.000007199288,0.00005159313,0.007963478,0.00005694873,0.00003118326,1.515856e-7,0.00003430519,0.0000012044,0.9889382,0.0009460829,0.001330477,0.0006391371],"study_design_scores_gemma":[0.00168166,0.0002764683,0.7592636,0.00009481698,0.0003373541,0.000007590512,0.0009675815,0.00008166074,0.1651719,0.009069426,0.06253317,0.0005148732],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9927862,0.003579515,0.00006266186,0.00003413262,0.0001161161,0.00009901377,0.002951473,0.000002536337,0.0003683541],"genre_scores_gemma":[0.9950731,0.0000730934,0.0002563502,0.00001251108,0.00000952909,0.00000191436,0.003983438,0.000003921558,0.000586177],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8237664,"threshold_uncertainty_score":0.2534073,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03962064740010846,"score_gpt":0.3130132495260148,"score_spread":0.2733926021259064,"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."}}