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

Table 5

The nine MDDLogo-clustered subgroups and their performances of fivefold cross-validations from 3325 acetylation sites in nonhistone dataset.

MDDLogo cluster	Frequency plot of sequence logo	Number of positive data	Sn	Sp	Acc	MCC	Reference (PMID)
nH1		817	0.847	0.858	0.855	0.655	22902405
nH2		285	0.860	0.894	0.885	0.715	22902405
nH3		570	0.884	0.899	0.895	0.742	22902405
nH4		495	0.911	0.948	0.939	0.841	19608861
nH5		137	0.934	0.964	0.956	0.885	—
nH6		275	0.891	0.898	0.896	0.746	22902405
nH7		141	0.894	0.929	0.920	0.796	—
nH8		121	0.884	0.920	0.911	0.775	22902405
nH9		484	0.771	0.794	0.788	0.514	—
All data		3325	0.863	0.885	0.880	0.706	—
Combined MDDLogo-clustered motifs		3325	0.863	0.885	0.880	0.706	—

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