VIRsiRNApred: a web server for predicting inhibition efficacy of siRNAs targeting human viruses

Table 2 Evaluation of performance of predictive models on validation dataset of 345 viral siRNAs(V ³⁴⁵ )

Predictive model no.	siRNA features		No. of siRNA features	Pearson correlation coefficient* on validation (V³⁴⁵) dataset# during 10-fold cross validation
Predictive model no.	siRNA features		No. of siRNA features	SVM	ANN	KNN	REP Tree
1		Mononucleotide frequency	4	0.16	0.08	0.09	0.08
2		Dinucleotide frequency	16	0.30	0.23	0.22	0.24
3		Trinucleotide frequency	64	0.39	0.25	0.24	0.26
4		Tetranucleotide frequency	256	0.40	0.26	0.27	0.28
5		Pentanucleotide frequency	1024	0.42	0.27	0.28	0.29
6		Binary	76	0.03	0.02	0.02	0.01
7		Thermodynamic features	21	0.19	0.15	0.18	0.15
8		Secondary structure	28	0.02	0.02	0.02	0.02
9		1 + 2 + 3 + 4 + 5	1364	0.48	0.32	0.34	0.30
10		6 + 9	1440	0.48	0.32	0.34	0.32
11		6 + 7 + 9	1461	0.50	0.33	0.35	0.31
12		6 + 7 + 8 + 9	1489	0.45	0.32	0.33	0.30

*Pearson Correlation Coefficient (PCC) is the correlation between experimental and predicted viral siRNA efficacy.
#V³⁴⁵ is the validation dataset of experimental viral siRNA not used in training. Predictive Models 1-8 were developed on individual siRNA features while models 9-12 were based on hybrid siRNA features.

ISSN: 1479-5876