{"id":"W6902328402","doi":"10.6084/m9.figshare.28059562.v1","title":"Additional file 5 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":[0.00001676249,0.00007499628,0.0001191189,0.00004131512,0.00001949323,0.00001105777,0.00008453256,0.00007080791,0.7232543],"category_scores_gemma":[0.001113527,0.00006850247,0.0001330428,0.0001121164,0.00001868129,0.000008816736,0.00004503885,0.00003078388,0.00005425258],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007559851,"about_ca_system_score_gemma":0.00008597153,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008477808,"about_ca_topic_score_gemma":0.00001806514,"domain_scores_codex":[0.9994368,0.00001419308,0.0001671296,0.0001973609,0.00009019999,0.00009434716],"domain_scores_gemma":[0.9993886,0.0002523067,0.00008783918,0.00009766789,0.0001476046,0.0000259393],"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":[0.00002806572,0.00001221735,7.822057e-7,0.0001713076,0.0000822591,0.000002216092,0.00002612315,0.0001518518,0.05944,0.000002366404,0.9387707,0.001312185],"study_design_scores_gemma":[0.0002863416,0.0003472291,0.00022516,0.002529534,0.00004616378,0.00001034661,0.0002188213,0.01772123,0.3372869,0.0001721744,0.6409162,0.0002398952],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0006925197,0.0002455896,0.00006132072,0.00002705757,0.00001585242,0.0001349593,0.9981052,0.000007866818,0.0007096549],"genre_scores_gemma":[0.1132083,0.000004790394,0.0003269651,0.00003482196,0.0001222696,0.0004400508,0.8844562,0.000008836014,0.001397855],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.7232,"threshold_uncertainty_score":0.2793449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02330589334251935,"score_gpt":0.2764935940420526,"score_spread":0.2531877006995332,"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."}}