. 2014 Jan 16;10(1):e1003440. doi: 10.1371/journal.pcbi.1003440

Table 3. Performance of consensus classifiers with PMD-UNIPROT and MMP datasets.

Performance metrics^a	PMD-UNIPROT			MMP
	CONDEL	Meta-SNP	PredictSNP	CONDEL	Meta-SNP	PredictSNP
Percent of evaluated mutations	100.0	100.0	100.0	100.0	99.7	100.0
Accuracy^b	0.562	0.670	0.679	0.640	0.673	0.708
Matthews correlation coefficient^b	0.202	0.343	0.366	0.349	0.351	0.433
Area under the receiver operating characteristics curve^b	0.755	0.709	0.732	0.770	0.730	0.780

– detailed evaluation is available in Table S12;

– these metrics were calculated with normalized numbers.