ViralPhos: incorporating a recursively statistical method to predict phosphorylation sites on virus proteins

BMC Bioinformatics

Table 1 Data statistics of training set and independent testing set.

	Data set		pSer	pThr	pTyr
Training set	virPTM	Positive data	233	54	14
		Negative data	2588	1170	65
		Balanced negative data	233	54	14
Independent testing set	dbPTM	Positive data	42	12	1
		Negative data	679	186	11
		Balanced negative data	42	12	1
	UniProtKB	Positive data	24	10	-
		Negative data	217	159	-
		Balanced negative data	24	10	-
	Phospho.ELM	Positive data	2	-	2
		Negative data	67	-	16
		Balanced negative data	2	-	2
	Combined non-redundant dataset	Positive data	42	12	2
		Negative data	352	106	16
		Balanced negative data	42	12	2

ISSN: 1471-2105