Comparison of Predictive In Silico Tools on Missense Variants in GJB2, GJB6, and GJB3 Genes Associated with Autosomal Recessive Deafness 1A (DFNB1A)

Table 2

Performance of in silico tools.

in silico Tools	Accuracy	Sensitivity	Specificity	PPV	NPV
SIFT	89%	67%	100%	100%	86%
MutationAssessor	78%	100%	67%	60%	100%
FATHMM	33%	100%	0%	33%	0%
Polyphen-2	78%	67%	83%	67%	50%
MutationTaster	56%	100%	33%	43%	33%
PROVEAN	89%	67%	100%	100%	86%
Align GVGD	44%	33%	33%	33%	67%
MutPred	67%	33%	83%	50%	71%
CONDEL	67%	100%	50%	50%	100%

Note. Accuracy (Aс) - the proportion of the correct test results (that is the sum of true positive and true negative results) among all the patients examined. In our case, this is the proportion of correct estimates of pathogenic and benign variants; Sensitivity (Se) - the ability of the diagnostic method to give the correct result which is defined as the proportion of true positive results among all performed tests. In our case, this is the proportion of true positive results, that is, the correct identification of pathogenic variants; Specificity (Sp) - the ability of the diagnostic method not to give false positive results in the absence of disease, which is defined as the proportion of true negative results among healthy individuals in studied group. In our case, this is a share of true negative results, that is, a correct identification of benign variants; Positive predictive values (PPV) - prediction of pathogenic variants; Negative predictive values (NPV) - prediction of benign variants.