{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0007394405,0.00174783,0.001458978,0.001667015,0.0008043465,0.002052603,0.002606928,0.001586497,0.7192292],"category_scores_gemma":[0.009332094,0.0008099461,0.001037567,0.001963908,0.0003763413,0.001310256,0.001214153,0.001242996,0.1744577],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001181178,"about_ca_system_score_gemma":0.001062864,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004350316,"about_ca_topic_score_gemma":0.008306519,"domain_scores_codex":[0.9995936,0.00005706602,0.00003218801,0.0001162087,0.0001311727,0.00006979744],"domain_scores_gemma":[0.9970944,0.002110425,0.000108983,0.0002561239,0.0002999732,0.0001300411],"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":[0.0003872154,0.0001105251,0.001617958,0.002047576,0.00007996106,0.0001485975,0.000102184,0.005210741,0.001136069,0.003542471,0.9703198,0.01529694],"study_design_scores_gemma":[0.003134502,0.0003221258,0.009609085,0.001225683,0.0002376562,0.0006654406,0.0003179302,0.07305753,0.01038924,0.06032874,0.8403949,0.0003172092],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001236233,0.00007221459,0.007350567,0.0001675155,0.0001125081,0.0001025104,0.9675832,0.01960675,0.003768444],"genre_scores_gemma":[0.0318407,0.0001898555,0.03371597,0.0004713596,0.0001123523,0.001314221,0.897848,0.02251874,0.01198886],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.7192292,"threshold_uncertainty_score":0.4004854,"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."}}