{"id":"W3127575774","doi":"10.1101/2021.01.31.429071","title":"A topological characterization of DNA sequences based on chaos geometry and persistent homology","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Fractal and DNA sequence analysis","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Topological data analysis; Persistent homology; Topology (electrical circuits); Homology (biology); DNA sequencing; Genome; Computational biology; DNA; Genetics; Biology; Mathematics; Algorithm; Gene; Combinatorics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002390447,0.0003172553,0.0004477163,0.0001724256,0.00007443249,0.00005987864,0.000247717,0.0005941236,0.00004767245],"category_scores_gemma":[0.0001766097,0.0003026181,0.0002354141,0.0002400249,0.0002572028,0.000007411873,0.0002710089,0.0002512809,0.0000026038],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003916281,"about_ca_system_score_gemma":0.0002758023,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003564778,"about_ca_topic_score_gemma":0.000001727954,"domain_scores_codex":[0.9982437,0.0001402458,0.0003368414,0.0008066551,0.0002025733,0.0002699328],"domain_scores_gemma":[0.9986129,0.00001802439,0.000303186,0.0006533882,0.0002832412,0.0001292644],"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":[0.00004272056,0.000125779,0.006272346,0.0001515334,0.0001396089,0.00002873446,0.000003051213,0.00007813584,0.993107,0.0000424554,0.000005333367,0.00000332154],"study_design_scores_gemma":[0.0002354409,0.000360794,0.06017441,0.0001325283,0.0001373207,6.786277e-8,0.000009428132,0.0005497776,0.937745,6.616646e-7,0.0002588319,0.0003957036],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9975806,0.0008564625,0.0007044227,0.0002885018,0.0002158159,0.0001916536,0.0001313531,0.00002408845,0.000007100086],"genre_scores_gemma":[0.9980806,0.000576851,0.0006101074,0.0004597394,0.0001723902,0.00005459031,0.00001218511,0.00002804613,0.000005479454],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05536195,"threshold_uncertainty_score":0.9999426,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.011044737686871,"score_gpt":0.2116497087783826,"score_spread":0.2006049710915116,"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."}}