Predicting Interactions between Virus and Host Proteins Using Repeat Patterns and Composition of Amino Acids

Abstract

Previous methods for predicting protein-protein interactions (PPIs) were mainly focused on PPIs within a single species, but PPIs across different species have recently emerged as an important issue in some areas such as viral infection. The primary focus of this study is to predict PPIs between virus and its targeted host, which are involved in viral infection. We developed a general method that predicts interactions between virus and host proteins using the repeat patterns and composition of amino acids. In independent testing of the method with PPIs of new viruses and hosts, it showed a high performance comparable to the best performance of other methods for single virus-host PPIs. In comparison of our method with others using same datasets, our method outperformed the others. The repeat patterns and composition of amino acids are simple, yet powerful features for predicting virus-host PPIs. The method developed in this study will help in finding new virus-host PPIs for which little information is available.

Figure 1

Example of computing feature 1 (F1) of amino acid repeats. F1 is the sum of squared length of single amino acid repeats (SARs) in the whole protein sequence. In this example, F1 (repeats of S) = 1² + 3² + 3² = 19, F1 (repeats of W) = 4² + 2² = 20, and F1 (repeats of R) = 1² + 6² = 37.

Figure 2

Example of computing feature 2 (F2) of amino acid repeats. F2 is the maximum value of the sum of squared length of single amino acid repeats in a window of size six. The maximum repeat size of amino acid S is 3, which is observed in the windows starting at 4, 5, 6, 7, 13, 14, and 15. So, F2 (repeats of S) = 3² = 9. The maximum repeat size of amino acid W is 4, observed in the windows starting at 1 and 2. F2 (repeats of W) = 4² = 16. The maximum repeat size of amino acid R is 6, observed in the window starting at 10. F2 (repeats of R) = 6² = 36.

Figure 3

Values of feature 2 for windows of six and seven residues. With a window of size 6, different patterns of single amino acid repeats lead to 15 different values for feature 2. With a window of size 7, different patterns of single amino acid repeats can lead to a same value for feature 2, as shown in this example.

Figure 4

ROC curves of independent testing of the SVM model on PPIs of new viruses with human. The SVM model showed the largest area under the ROC curve (AUC) of 0.970 for the PPIs of the SARS virus with human.

Figure 5

ROC curves of independent testing of the SVM model on PPIs of new hosts with viruses. The SVM model showed the largest area under the ROC curve (AUC) of 0.959 for the PPIs of E. coli K-12 with viruses.