Fig. 3From: A benchmarking framework for the accurate and cost-effective detection of clinically-relevant structural variants for cancer target identification and diagnosisConstruction of the random-forest decision model for prediction of bona fide SVs. A Workflow of the random-forest decision model construction. B The distribution of cancer types in Cohort 1 (Up) and Cohort 2 (Down). C Confusion matrixes in the training set (left), testing set1 (middle), and testing set2 (right). D Performance of random-forest decision model in the training set (left), testing set1 (middle), and testing set2 (right)Back to article page