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

Table 3 Comparison of VIRsiRNApred with existing siRNA efficacy prediction algorithms developed using heterogeneous siRNA dataset

S.no	Reference	Url	Technique	siRNA data set	Pearson correlation coefficient*
S.no	Reference	Url	Technique	siRNA data set	Train¹	Val²	V³⁴⁵#
1	[34]	NA	GPBoost, SVM	581	0.46	0.40	Server not available
2	[37]	NA	ANN	653	0.55	0.50	Server not available
3	[32]	http://biodev.extra.cea.fr/DSIR/DSIR.html	linear	653	0.48	0.44	Server not working
4	[36]	NA	linear	526	0.55	0.52	Server not available
5	[38]	http://www.med.nagoya-u.ac.jp/neurogenetics/i_Score/i_score.html	linear	419	0.51	0.44	Server not working
6	[46]	http://www.imtech.res.in/raghava/sirnapred	SVM	581	0.56	0.47	0.10
7	VIRsiRNApred	http://crdd.osdd.net/servers/virsirnapred/	SVM	1380	0.58	0.55	0.55

*Pearson Correlation Coefficient (PCC) is the correlation between experimental and predicted viral siRNA efficacy.
¹Performance on n-fold training dataset of the study.
²Performance on validation data set of the study.
#V³⁴⁵ is the validation dataset of experimental viral siRNA. Algorithms from S.No. 1-6 used mammalian heterogeneous siRNA datasets while S.No. 7 used experimental viral siRNA dataset.

ISSN: 1479-5876