Prioritization of pathogenic mutations in the protein kinase superfamily

BMC Genomics

Table 5 Summary of the performance of other state-of-the-art classifiers of mutations, either general or kinase-specific

Method	Scope	Accuracy (%)	Precision (%)	Recall (%)	MCC
KinMut	Kinase†	83.3	60.0	75.2	0.6
SNPs&GO [18]	Kinase†	82.3	62.8	77.5	0.6
Torkamani [19]	Kinase	77.0	-	-	0.5
MutationAssessor [9]	Kinase†	53.8	41.6	95.6	0.5
SNAP [16]	Kinase†	49.4	34.0	93.1	0.4
SIFT [7]	Kinase†	77.6	37.8	27.9	0.2
SNPs&GO [18]	Genome-wide	82.0	83.0	78.0	0.6
MutationAssessor [9]	Genome-wide	79.0	-	-	-
SNAP [16]	Genome-wide	78.2	76.7	80.2	-
SIFT [7]	Genome-wide	68.3	66.1	56.5	0.3

Summary of the performance of other state-of-the-art classifiers of mutations, either general or kinase-specific. Performance was measured in terms of overall accuracy recall and the Matthews correlation coefficient. General methods with which the prediction corresponds to our dataset are marked with †. The remaining results for the classifiers displayed here were taken directly from their original publications

ISSN: 1471-2164