Skip to main content

Table 1 Performance of classifiers in predicting active OPEN target sites

From: Predicting success of oligomerized pool engineering (OPEN) for zinc finger target site sequences

Classifier Target site
encoding
ROC AUC Correlation
Coefficient
Accuracy % Specificity+% Sensitivity+%
Naïve Bayes ZiFOpT
(Sequence Identity)
0.89 0.61 87 90 94
  Base Counts 0.79 0.57 87 89 94
  Positional Base Counts 0.84 0.59 87 88 97
SVM Sequence Identity 0.76 0.48 84 86 95
  Base Counts 0.78 0.54 85 89 92
  Positional Base Counts 0.84 0.63 88 90 95