Skip to main content
. 2020 Jul 16;10:11770. doi: 10.1038/s41598-020-68251-4

Figure 3.

Figure 3

Results comparing the predictive performance of RPmirDIP and RPmirDIP* models against the original mirDIP predictions. (A) and (B) depict the ROC and PR curves, respectively. (C) illustrates the change in predictive performance as a function of the size of the training set. (D) is a kernel density estimation plot of the pair-wise difference in scores between RPmirDIP and mirDIP.