{"id":"W6958649389","doi":"10.6084/m9.figshare.28059556","title":"Additional file 3 of CGRclust: Chaos Game Representation for twin contrastive clustering of unlabelled DNA sequences","year":2024,"lang":"en","type":"article","venue":"Figshare","topic":"Genetic and Environmental Crop Studies","field":"Agricultural and Biological Sciences","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":[0.00001139092,0.00006023024,0.0000998666,0.000005772086,0.00003985454,0.000009390919,0.00006404099,0.00003130443,0.9283744],"category_scores_gemma":[0.0002847002,0.00002610618,0.00006655398,0.00008314687,0.00002030339,0.00003991684,0.00005414494,0.00002370198,0.0001142341],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009567883,"about_ca_system_score_gemma":0.000005052471,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001114449,"about_ca_topic_score_gemma":0.00003635858,"domain_scores_codex":[0.9995467,0.00001064502,0.000122874,0.0001368417,0.00009618598,0.00008675872],"domain_scores_gemma":[0.9988884,0.0009925911,0.0000543985,0.00001741785,0.00002828904,0.00001891903],"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":[0.000009317294,0.00001174526,0.000004696433,0.00006260913,0.00001835989,7.207901e-7,0.00006310871,0.00002686657,0.007261999,0.000001240298,0.9847688,0.007770547],"study_design_scores_gemma":[0.0001724118,0.0005263656,0.0665886,0.003042088,0.00002075993,0.000005629878,0.00167049,0.003712177,0.01370838,0.0002823022,0.910022,0.0002487431],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00388257,0.0001521993,7.803186e-7,0.00006884276,0.00001643501,0.0001719005,0.9949935,0.00001338523,0.0007003746],"genre_scores_gemma":[0.2629439,0.000005077507,0.0001484855,0.00002291827,0.0001042461,0.0005350602,0.7350888,8.940651e-7,0.001150637],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.9282601,"threshold_uncertainty_score":0.1468287,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04253279661907856,"score_gpt":0.236589258884836,"score_spread":0.1940564622657574,"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."}}