{"id":"W6921191962","doi":"10.6084/m9.figshare.25067253.v1","title":"Additional file 1 of Anchor Clustering for million-scale immune repertoire sequencing data","year":2024,"lang":"en","type":"dataset","venue":"Figshare","topic":"Mathematics Education and Programs","field":"Mathematics","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Cluster analysis; Repertoire; DNA sequencing; Key (lock); Data file","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":["metaresearch","metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0001058702,0.0002832119,0.0004554326,0.0001274542,0.00009061815,0.00009696838,0.0009545852,0.0002838613,0.9935049],"category_scores_gemma":[0.01029409,0.0002647481,0.0001827116,0.0001774881,0.00001212968,0.00009891918,0.0006008515,0.00030367,0.005137334],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007794463,"about_ca_system_score_gemma":0.00042927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004652551,"about_ca_topic_score_gemma":0.00007025343,"domain_scores_codex":[0.9983301,0.00002680704,0.0006326081,0.0004660328,0.0003052784,0.0002391828],"domain_scores_gemma":[0.9935132,0.004040799,0.0005095108,0.001654883,0.0002063831,0.0000751839],"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":[0.000002905189,0.00005108617,4.773443e-10,0.01242561,0.00006455189,0.000003017423,0.00005322236,1.872642e-7,4.047006e-7,0.000002002038,0.9867358,0.0006612138],"study_design_scores_gemma":[0.00005904144,0.00002537363,6.818925e-8,0.02468611,0.00005976409,0.0000242892,0.00009165704,0.0008383133,0.000007977414,0.001343404,0.9725996,0.0002643717],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[9.335189e-9,0.0001746645,0.000003303049,0.00002878012,0.0001512906,0.000661747,0.9984754,0.00006985478,0.0004349048],"genre_scores_gemma":[1.597408e-8,0.000004692189,0.02041039,0.00004224887,0.0005040148,0.003270537,0.9734312,0.00005815439,0.002278798],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.9883676,"threshold_uncertainty_score":0.9999804,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1923862650353638,"score_gpt":0.3682997128319365,"score_spread":0.1759134477965727,"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."}}