An Intelligent System for Identifying Acetylated Lysine on Histones and Nonhistone Proteins

Table 4

The five MDDLogo-clustered subgroups and their performances of fivefold cross-validations from 200 acetylation sites in histone dataset.

MDDLogo cluster	Frequency plot of sequence logo	Number of positive data	Sn	Sp	Acc	MCC	Reference (PMID)
H1		43	0.930	0.944	0.942	0.812	10485713
H2		29	0.931	0.938	0.937	0.799	22902405
H3		28	0.929	0.936	0.935	0.793	16916647
H4		22	0.909	0.918	0.917	0.745	—
H5		78	0.897	0.910	0.908	0.722	—
All data		200	0.915	0.926	0.924	0.764	—
Combined MDDLogo-clustered motifs		200	0.915	0.926	0.924	0.764	—

Sn, sensitivity; Sp, specificity; Acc, accuracy; MCC, Matthew’s correlation coefficient.