{"id":"W3133739977","doi":"10.1101/2021.03.10.434824","title":"A topological approach to DNA similarity analysis from 5-dimensional representation","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":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Topological data analysis; Persistent homology; DNA sequencing; Topology (electrical circuits); DNA; Representation (politics); Similarity (geometry); Homology (biology); Mathematics; Computational biology; Computer science; Genetics; Biology; Algorithm; Combinatorics; Artificial intelligence; Gene","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":[0.0003038996,0.0004575798,0.0006976576,0.0002379386,0.0001386964,0.0002555149,0.0005052185,0.0007663895,0.0001032826],"category_scores_gemma":[0.0003675347,0.0004618276,0.0006072009,0.00102211,0.0001069319,0.00001505097,0.0009816999,0.0004198531,0.00002032635],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008296793,"about_ca_system_score_gemma":0.0002985823,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007250429,"about_ca_topic_score_gemma":0.00003798691,"domain_scores_codex":[0.9966758,0.000237098,0.0004872984,0.00175738,0.0004329826,0.0004093654],"domain_scores_gemma":[0.9973137,0.00002815749,0.0002492926,0.001577783,0.0004951773,0.0003358427],"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":[0.00005616302,0.0002793209,0.01295697,0.00002409864,0.001745277,0.00003448807,0.000006638688,0.004335726,0.9802063,0.0000312624,0.000322404,0.000001400864],"study_design_scores_gemma":[0.000299882,0.0000679703,0.17679,0.00003763035,0.001651679,1.894965e-8,0.00002285541,0.002467439,0.8167573,0.000005697604,0.0009189862,0.0009806051],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9855458,0.0007418452,0.0125546,0.0002933561,0.0002278593,0.0002743387,0.0002677758,0.00006409641,0.00003036346],"genre_scores_gemma":[0.9801729,0.00008555414,0.01796047,0.001061603,0.0004749152,0.0001298047,0.00006244572,0.00004111558,0.00001124904],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.163833,"threshold_uncertainty_score":0.9997833,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01909908138525299,"score_gpt":0.2461632364918969,"score_spread":0.2270641551066439,"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."}}