{"id":"W2109550992","doi":"10.1093/bioinformatics/bth946","title":"Into the heart of darkness: large-scale clustering of human non-coding DNA","year":2004,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genome Rearrangement Algorithms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":69,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Human genome; Biology; Genome; Computational biology; Conserved sequence; Cluster analysis; Genetics; Gene; Coding region; Computer science; Peptide sequence; Artificial intelligence","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":[0.0002460653,0.00009853231,0.0001434922,0.00003985485,0.00007839649,0.00001096494,0.000214727,0.0000709485,0.000007744926],"category_scores_gemma":[0.00001548003,0.0000758222,0.00008072639,0.00008777041,0.00007506261,0.000007061411,0.0001890763,0.00004849183,0.000006538466],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001309059,"about_ca_system_score_gemma":0.00003678253,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001830215,"about_ca_topic_score_gemma":0.0000410174,"domain_scores_codex":[0.9992583,0.000008568634,0.0003474792,0.00007454331,0.0001443824,0.0001666739],"domain_scores_gemma":[0.9994085,0.000004318018,0.0001508689,0.0003386747,0.00006562649,0.00003203528],"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":[0.00001584702,0.00009260131,0.0009369761,0.0004692294,0.00007567384,3.219089e-7,0.007953673,0.0007837158,0.9880705,0.0001164336,0.0003058026,0.001179238],"study_design_scores_gemma":[0.001584209,0.0004768419,0.003288626,0.000128884,0.00004275891,0.000008969277,0.004928622,0.003826654,0.9790202,0.0001548264,0.006267874,0.0002715671],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9410316,0.0001294811,0.0550522,0.0001053655,0.0001154744,0.0002693583,0.00001378368,0.000008159042,0.003274516],"genre_scores_gemma":[0.9797614,0.00004144003,0.01993988,0.00007199179,0.00005734717,0.000006231495,0.00004243581,0.00001113345,0.00006809169],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03872978,"threshold_uncertainty_score":0.309194,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009885294791323542,"score_gpt":0.2488597369418888,"score_spread":0.2389744421505653,"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."}}