Skip to main content
. 2021 Jan 13;19:3. doi: 10.1186/s12915-020-00930-0

Table 3.

Comparison of cancer driver variant predictors on the DRGN benchmark data sets. The last column shows the AUC drop with respect the EXPDRV+ALL data set in Table 2

Method AUC AUPRC AUC drop
ParsSNP 59.4 52.0 27.5
CanDrA v+ 66.2 64.7 20.9
Chasm 3.1 76.6 74.1 14.5
CHASMplus 72.3 69.8 22.0
FATHMM 69.1 65.9 21.9
TransFIC 59.6 53.0 8.4
Condel 55.9 51.7 8.4
DEOGEN2 57.2 48.7 11.6
CADD 63.6 55.5 10.1
M-CAP 47.4 45.9 6.6
MetaSVM 55.7 49.2 11.1
Dummy 50.0 44.2 48.1