Skip to main content
. 2024 Jan 10;4(1):100687. doi: 10.1016/j.crmeth.2023.100687

Figure 3.

Figure 3

The performance of SIGMA

(A) The distribution of SIGMA scores for 20,047 pathogenic (red) and 20,148 benign (blue) variants in the training set. SIGMA scores were calculated using out-of-fold predictions.

(B and C) Receiver operating characteristic (ROC) curves of SIGMA scores in the training set (B) and test set (C).

(D) Comparison of area under the ROC curves (AUCs) of SIGMA and 16 individual variant effect predictors (VEPs) using the test set.

(E) Rank of correlation between individual VEPs and deep mutational scanning (DMS) measurements. Spearman’s correlation was calculated between functional scores from DMS experiments and prediction scores from SIGMA as well as 17 individual VEPs. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.