Skip to main content

Advertisement

Table 5 Performance of the SVM models using leave one out cross validation (LOOCV) method

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

Predictive model no. siRNA features No. of siRNA features Pearson correlation coefficient*
Training (T1380) Validation (V345)
1   Mononucleotide frequency 4 0.32 0.29
2 Dinucleotide frequency 16 0.36 0.32
3 Trinucleotide frequency 64 0.45 0.41
4 Tetranucleotide frequency 256 0.48 0.44
5 Pentanucleotide frequency 1024 0.52 0.48
6 Binary 76 0.26 0.14
7 Thermodynamic features 21 0.29 0.24
8 Secondary structure 28 0.10 0.06
9   1 + 2 + 3 + 4 + 5 1364 0.52 0.49
10 6 + 9 1440 0.54 0.51
11 6 + 7 + 9 1461 0.58 0.55
12   6 + 7 + 8 + 9 1489 0.58 0.54
  1. *Pearson Correlation Coefficient (PCC) is the correlation between experimental and predicted viral siRNA efficacy.
  2. # T1380 is the training dataset of experimental viral siRNA. Predictive Models 1-8 were developed on individual siRNA features while models 9-12 were based on hybrid siRNA features.