Prediction of mucin-type O-glycosylation sites in mammalian proteins using the composition of k-spaced amino acid pairs

Table 1 Prediction accuracy of O-glycosylation sites based on different encoding schemes^a

Site	Encoding scheme	Feature selection	Sn (%)	Sp (%)	Ac (%)	MCC
S	Binary^b	No selection	74.2 ± 1.7	81.9 ± 3.0	78.0 ± 1.9	0.567 ± 0.039
	Binary^c	No selection	76.5 ± 3.5	74.6 ± 3.6	75.6 ± 3.1	0.523 ± 0.060
	CKSAAP	No selection	77.9 ± 1.7	86.5 ± 3.0	82.2 ± 1.8	0.655 ± 0.037
	CKSAAP^c	No selection	79.0 ± 5.2	83.0 ± 2.4	81.0 ± 2.6	0.628 ± 0.050
	CKSAAP	CC	80.7 ± 3.3	85.6 ± 3.9	83.1 ± 2.8	0.671 ± 0.055
	CKSAAP	IE	82.1 ± 2.3	83.9 ± 3.8	83.0 ± 2.4	0.665 ± 0.048
T	Binary^b	No selection	74.8 ± 4.1	78.3 ± 1.7	76.6 ± 2.3	0.536 ± 0.045
	Binary^c	No selection	77.8 ± 3.4	76.6 ± 3.2	77.2 ± 2.4	0.548 ± 0.048
	CKSAAP	No selection	80.4 ± 2.2	82.3 ± 2.9	81.3 ± 2.3	0.631 ± 0.045
	CKSAAP^c	No selection	80.3 ± 1.9	85.7 ± 1.9	83.0 ± 1.8	0.666 ± 0.038
	CKSAAP	CC	80.3 ± 1.8	82.5 ± 2.3	81.4 ± 1.3	0.632 ± 0.026
	CKSAAP	IE	80.8 ± 1.5	81.9 ± 3.1	81.3 ± 2.2	0.631 ± 0.045

^a The SVM based prediction algorithm with the RBF kernel function. The CC-based feature selection resulted in the highest accuracy, and the corresponding values of Ac and MCC were represented in bold types. The corresponding measurement was represented as the average value ± standard deviation. ^b In this encoding scheme, the window size was optimally set as 41. ^cThe method was trained and tested on new negative site data sets where <40% identity was not required between positive and negative sites.

ISSN: 1471-2105