DVA: predicting the functional impact of single nucleotide missense variants

BMC Bioinformatics

Table 3 Performance of prediction methods using missense variants in the VariBench database

Order	Methods	Accuracy	Precision	Recall	F1-score
1	SIFT	0.625	0.631	0.547	0.582
2	MutationTaster	0.536	0.512	0.77	0.615
3	MutationAssessor	0.631	0.65	0.523	0.573
4	FATHMM-MKL_coding	0.603	0.574	0.694	0.626
5	PROVEAN	0.617	0.611	0.594	0.593
6	MetaSVM	0.734	0.738	0.703	0.717
7	MetaLR	0.742	0.753	0.699	0.722
8	DANN	0.628	0.627	0.584	0.598
9	CADD	0.633	0.616	0.632	0.623
10	MISTIC	0.739	0.75	0.696	0.718
11	REVEL	0.755	0.751	0.741	0.744
12	PrimateAI	0.621	0.602	0.645	0.62
13	ClinPred	0.692	0.691	0.656	0.67
14	M-CAP	0.746	0.764	0.688	0.722
15	DVA	0.785	0.797	0.744	0.768

ISSN: 1471-2105