Additional file 3 of Letter to the Editor: clinical utility of urine DNA for noninvasive detection and minimal residual disease monitoring in urothelial carcinoma

Additional file 3: Supplementary Table 1. Patient characteristics of the training cohort. Supplementary Table 2. Characteristics of the UTUC validation cohort. Supplementary Table 3. Characteristics of the utLIFE-UC MRD cohort. Supplementary Table 4. Mutation used as candidate markers. Supplementary Table 5. AUC of different Cutoff_1 value in the training set. Supplementary Table 6. The diagnostic performance comparing the CNV score model with all autosomes and with specific chromosomes. Supplementary Table 7. Performance of the ML models in modeling cohort. Supplementary Table 8. Performance of TCGA validation cohorts by 3 ML models.

附加文件3：补充表1。训练队列的患者临床特征。补充表2：上尿路尿路上皮癌（UTUC）验证队列的特征。补充表3：utLIFE-UC微小残留病（MRD）队列的特征。补充表4：用作候选标志物的突变。补充表5：训练集中不同Cutoff_1阈值对应的受试者工作特征曲线下面积（AUC）。补充表6：对比全常染色体与特定染色体的拷贝数变异（CNV）评分模型的诊断性能。补充表7：机器学习（ML）模型在建模队列中的性能表现。补充表8：3个机器学习（ML）模型对癌症基因组图谱（TCGA）验证队列的性能评估。