Exploiting structural and topological information to improve prediction of RNA-protein binding sites

BMC Bioinformatics

Table 2 Prediction performance for different patch types.

Classifier	Patch type	AUC	δ ₉₅	MCC	SN[%]	SP[%]
NB	sequential	0.77	0.016	0.30	73.6	66.8
NB	topological	0.78	0.016	0.31	75.4	65.7
NB	spatial	0.79	0.016	0.32	75.4	67.1
SVM	sequential	0.79	0.021	0.34	78.4	65.8
SVM	topological	0.80	0.021	0.34	79.9	64.4
SVM	spatial	0.80	0.021	0.36	80.0	65.6

Prediction performance for different patch types (sequential, topological, spatial) and classifiers (NB, SVM) on RB144 data set, tested by five-fold cross-validation. C-value for SVM was 1.0 and γ-value of RBF-kernel was 0.01. Cost factor was assigned as 5.7 to compensate for the unbalanced class distribution.

ISSN: 1471-2105