{"id":"W6939671747","doi":"10.6084/m9.figshare.28059565.v1","title":"Additional file 6 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":[0.00001676308,0.00007499714,0.0001191176,0.0000413143,0.0000194899,0.00001106309,0.00008453986,0.00007080439,0.721599],"category_scores_gemma":[0.001113576,0.00006850552,0.000133033,0.0001121226,0.00001868241,0.000008817498,0.00004504087,0.0000307862,0.00005437057],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007550251,"about_ca_system_score_gemma":0.00008598578,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008564527,"about_ca_topic_score_gemma":0.00001837634,"domain_scores_codex":[0.9994367,0.0000141937,0.0001671333,0.0001973644,0.00009020036,0.00009435671],"domain_scores_gemma":[0.9993886,0.000252346,0.00008784102,0.00009766908,0.0001476364,0.00002594061],"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":[0.00002823349,0.00001228792,7.811315e-7,0.0001715048,0.0000823556,0.000002217629,0.0000262292,0.0001499062,0.05942141,0.0000023894,0.93879,0.001312684],"study_design_scores_gemma":[0.0002849083,0.0003453067,0.00022238,0.002514192,0.00004590858,0.0000102814,0.000217947,0.01751804,0.333315,0.000172113,0.6451154,0.0002385948],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0006879842,0.000246364,0.00006146955,0.00002708263,0.0000158551,0.0001351557,0.9980996,0.000007866618,0.0007186847],"genre_scores_gemma":[0.113342,0.00000478594,0.0003270346,0.00003483973,0.0001223395,0.0004407949,0.884312,0.000008840868,0.001407358],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.7215446,"threshold_uncertainty_score":0.2793574,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02329892889764531,"score_gpt":0.2764528755006588,"score_spread":0.2531539466030135,"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."}}