{"id":"W4414404509","doi":"10.1101/2025.09.20.677550","title":"MicroNucML: A machine learning approach for micronuclei segmentation and the refinement of nuclei-micronuclei relationships","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Nuclear Physics and Applications","field":"Physics and Astronomy","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University Health Network","funders":"Princess Margaret Cancer Foundation","keywords":"Segmentation; Genome instability; Micronucleus test; Fidelity; Pattern recognition (psychology); Polygon (computer graphics); Object (grammar); Genome","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006558773,0.0003873114,0.0005008171,0.0001211703,0.0005381969,0.0001605708,0.0003947623,0.0001514113,0.00003165874],"category_scores_gemma":[0.00002285936,0.00035123,0.0002231416,0.0002181842,0.0002256072,0.00007605553,0.0004521139,0.0007188868,0.000004993063],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006819299,"about_ca_system_score_gemma":0.0001557085,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000190581,"about_ca_topic_score_gemma":8.573593e-7,"domain_scores_codex":[0.9981468,0.0001677403,0.0005702925,0.0006536772,0.0001603606,0.0003010841],"domain_scores_gemma":[0.9981441,0.0001689751,0.0006281993,0.0007056708,0.0002700067,0.00008308578],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002679923,0.0008389004,0.01281268,0.001382091,0.001332315,3.103801e-7,0.0003775551,0.002128193,0.6905946,0.2885925,0.001464334,0.0002085138],"study_design_scores_gemma":[0.02710352,0.0003312281,0.05544602,0.002108227,0.005395857,4.150222e-8,0.0007589544,0.1519693,0.6127105,0.00214822,0.1360197,0.006008315],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8196006,0.002092504,0.1633639,0.001628673,0.0004845137,0.00750361,0.003986713,0.0003336041,0.001005953],"genre_scores_gemma":[0.9653881,0.0000621108,0.03354172,0.00005509503,0.0002103511,0.0006038669,0.00001321918,0.00007734667,0.00004818169],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2864443,"threshold_uncertainty_score":0.999894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01383516248025878,"score_gpt":0.2211446446646928,"score_spread":0.207309482184434,"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."}}