DVA: predicting the functional impact of single nucleotide missense variants

BMC Bioinformatics

Table 2 Performance of prediction methods using missense variants in the ClinVar database

Order	Methods	Accuracy	Precision	Recall	F1-score
1	SIFT	0.756	0.803	0.748	0.773
2	MutationTaster	0.629	0.624	0.846	0.718
3	MutationAssessor	0.776	0.836	0.747	0.788
4	FATHMM-MKL_coding	0.735	0.743	0.807	0.772
5	PROVEAN	0.794	0.818	0.814	0.815
6	MetaSVM	0.799	0.845	0.785	0.813
7	MetaLR	0.774	0.806	0.79	0.795
8	DANN	0.741	0.75	0.809	0.776
9	CADD	0.798	0.806	0.843	0.824
10	MISTIC	0.805	0.854	0.788	0.819
11	REVEL	0.85	0.889	0.837	0.862
12	PrimateAI	0.731	0.751	0.781	0.764
13	ClinPred	0.909	0.942	0.892	0.916
14	M-CAP	0.795	0.826	0.804	0.814
15	DVA	0.929	0.946	0.926	0.935

ISSN: 1471-2105