{"id":"W4405514853","doi":"10.1186/s12864-024-11135-y","title":"CGRclust: Chaos Game Representation for twin contrastive clustering of unlabelled DNA sequences","year":2024,"lang":"en","type":"article","venue":"BMC Genomics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University; University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Cluster analysis; Biology; DNA sequencing; Unsupervised learning; Computational biology; Sequence (biology); Genome; Pattern recognition (psychology); Artificial intelligence; DNA microarray; Sequence alignment; Computer science; DNA; Genetics; Peptide sequence; Gene","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0005946321,0.001146872,0.0007339959,0.001328793,0.0005578811,0.0009974788,0.00207543,0.001215236,0.004089254],"category_scores_gemma":[0.002272191,0.000380683,0.001133622,0.0008556491,0.0007170851,0.00118767,0.001182563,0.001277022,0.001285427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001543119,"about_ca_system_score_gemma":0.001450087,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008221707,"about_ca_topic_score_gemma":0.01149405,"domain_scores_codex":[0.9995437,0.00007658439,0.00002669661,0.0001589594,0.0001380161,0.00005604301],"domain_scores_gemma":[0.9994216,0.0001565404,0.00009032903,0.0001167756,0.0001598732,0.00005496501],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005197168,0.0001631495,0.003833857,0.0003263275,0.000169464,0.000307364,0.0003116956,0.4782205,0.02925654,0.02889264,0.01902732,0.4389714],"study_design_scores_gemma":[0.00000871508,0.00002343245,0.0001742916,0.000006455722,0.00000556231,0.00002518394,0.00001282252,0.9899758,0.003130863,0.005092101,0.001535515,0.000009324768],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03767069,0.0002364434,0.9488631,0.0002847546,0.00009536475,0.0001781398,0.001081911,0.008832752,0.002756828],"genre_scores_gemma":[0.3187005,0.0001797119,0.6708304,0.0003784263,0.00005988852,0.0005261226,0.004031587,0.0009491247,0.004344197],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008221707,"threshold_uncertainty_score":0.01634771,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02937545361735201,"score_gpt":0.2847074385826034,"score_spread":0.2553319849652514,"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."}}