FIGURE 1.

ΔCt₃₇₅ analysis of the training set and validation set samples. (A) Scatter plot for ΔCt₃₇₅ analysis (Ct_u6–Ct₃₇₅) of the training set samples: color-code divides the low-grade NE (red) and non-NE (blue). The quantity ΔCt₃₇₅ is able to discriminate low-grade NE from non-NE with 92.6% of sensitivity and 90.4% of specificity. (B) Probability density functions of ΔCt₃₇₅ relative expression for low-grade NE (red) and non-NE (blue). (C) Scatter plot for ΔCt₃₇₅ analysis of the validation set samples: color-code divides the low-grade NE (red) and non-NE (blue). Empty-crossed circles represent technical outliers. (D) ROC curve for ΔCt₃₇₅ which results in an AUC = 0.88.