Skip to main content

Advertisement

Table 2 Accuracy of three different machine learning prediction algorithms (J48 Decision Tree, Naïve Bayes and SVM with SMO training) using the frequencies of all possible short tripeptide binary segments.a

From: Use of machine learning algorithms to classify binary protein sequences as highly-designable or poorly-designable

  J48 Naïve Bayes SMO
a) Sequences folding to the top 10% and the bottom 10% of designable conformations for the hexagon 89.7% correct 78.8% correct 91.0% correct
  AUC 0.95 AUC 0.92 AUC 0.91
  Sens: 0.91 Sens: 0.85 Sens: 0.84
  Spec: 0.90 Spec: 0.77 Spec: 0.91
b) Sequences folding to the top 10% and the bottom 10% of designable conformations for the triangle 67.8% correct 56.7% correct 57.8% correct
  AUC 0.69 AUC 0.61 AUC 0.58
  Sens: 0.68 Sens: 0.58 Sens: 0.64
  Spec: 0.68 Spec: 0.57 Spec: 0.57
  1. a We compare random subsets of sequences corresponding to the top 10% and the bottom 10% of designabile structures for the a) hexagon, and b) triangle. Prediction accuracy and area under the curve (AUC), sensitivity (Sens) and specificity (Spec) for each method are given.