{"id":"W6921226785","doi":"10.6084/m9.figshare.28059568.v1","title":"Additional file 7 of CGRclust: Chaos Game Representation for twin contrastive clustering of unlabelled DNA sequences","year":2024,"lang":"en","type":"article","venue":"Figshare","topic":"Fractal and DNA sequence analysis","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University; University of Waterloo","funders":"","keywords":"Cluster analysis; Representation (politics); CHAOS (operating system); Pattern recognition (psychology); Feature (linguistics)","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00001676217,0.00007500161,0.0001191203,0.00004131794,0.00001949368,0.00001106345,0.00008454176,0.00007080613,0.7204324],"category_scores_gemma":[0.001102037,0.00006850911,0.0001330355,0.0001121173,0.00001868405,0.00000881782,0.0000450423,0.00003078652,0.00005436543],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007561146,"about_ca_system_score_gemma":0.0000859989,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008482997,"about_ca_topic_score_gemma":0.0000180732,"domain_scores_codex":[0.9994367,0.00001418976,0.0001671325,0.0001973815,0.0000902012,0.00009436419],"domain_scores_gemma":[0.9993888,0.0002521396,0.00008783615,0.00009767235,0.000147627,0.00002594184],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0000281311,0.00001224809,7.834573e-7,0.0001732252,0.0000824747,0.00000222183,0.00002622319,0.0001524697,0.05972886,0.0000023662,0.9384683,0.001322665],"study_design_scores_gemma":[0.0002846835,0.0003457683,0.0002193465,0.002539084,0.0000458963,0.00001028432,0.0002174033,0.01773429,0.3365869,0.0001719973,0.6416059,0.0002385392],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0006922961,0.000245778,0.00006140277,0.00002706694,0.00001585505,0.0001349421,0.9981051,0.000007866943,0.0007096734],"genre_scores_gemma":[0.1127917,0.000004796937,0.0003267293,0.00003480511,0.0001222242,0.0004394359,0.884874,0.000008832853,0.001397421],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.720378,"threshold_uncertainty_score":0.2798231,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02333777291267821,"score_gpt":0.2765415422726114,"score_spread":0.2532037693599332,"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."}}