Fig. 5From: Identification of Hürthle cell cancers: solving a clinical challenge with genomic sequencing and a trio of machine learning algorithmsNeoplasm Classifier Cross Validation Performance. a Samples used in the Classifier development. Hürthle and Neoplasm labels are defined by cytopathology. Note that all samples are Hürthle positive. b Volcano plots of differential expression. Fold-change (log2 scale) is plotted on the x-axis, and FDR-adjusted p-values are plotted on the y-axis. Mitochondrial genes are shown in purple. c Neoplasm Index Score. Red dashed line indicates the cut-off for NI+ vs. NI-. Cyan triangles indicate genome-wide LOH positive. b ROC curve showing classifier performance. The red-dashed lines indicate performance at the selected cutoffBack to article page