{"id":"W6958508691","doi":"10.6084/m9.figshare.28059559","title":"Additional file 4 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":[0.00001675413,0.00007498892,0.0001190987,0.00004131677,0.00001949233,0.00001106261,0.00008456407,0.0000708045,0.7233079],"category_scores_gemma":[0.001109111,0.00006850121,0.000133025,0.0001120806,0.0000186812,0.00000881675,0.00004503659,0.00003078029,0.00005461648],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007558213,"about_ca_system_score_gemma":0.00008587588,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008402181,"about_ca_topic_score_gemma":0.00001800104,"domain_scores_codex":[0.9994368,0.00001418124,0.0001671349,0.0001973359,0.0000901947,0.0000943495],"domain_scores_gemma":[0.9993895,0.0002526236,0.00008784304,0.00009768517,0.0001464267,0.00002593664],"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":[0.0000285426,0.0000122327,7.859343e-7,0.000171488,0.00008236813,0.000002219149,0.00002617181,0.0001502076,0.0591045,0.000002368303,0.9391072,0.00131191],"study_design_scores_gemma":[0.0002858292,0.0003510372,0.0002210073,0.002521696,0.00004603746,0.00001032098,0.0002181915,0.01780297,0.3341883,0.0001726733,0.6439426,0.0002392922],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0006891108,0.0002458426,0.00006149918,0.00002690975,0.00001584877,0.0001349648,0.9981072,0.000007864875,0.0007107519],"genre_scores_gemma":[0.1140419,0.000004784395,0.0003265528,0.00003475467,0.0001220949,0.0004395005,0.8836358,0.000008823177,0.001385779],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.7232533,"threshold_uncertainty_score":0.2793398,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02348467385559504,"score_gpt":0.2767670873380569,"score_spread":0.2532824134824619,"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."}}