Algorithm

After collecting the data, support vector machine algorithm was used to predict protein-protein interaction.

Support Vector Machine (SVM)

SVMs are universal approximators based on statistical and optimization theory. The SVM is particularly attractive to biological sequence analysis due to its ability to handle noise, large dataset and large input spaces. In the present study, we have used LIBSVM to predict the protein-protein interaction. The software allows the users to define a number of parameters and also enables a choice of inbuilt kernel function, like linear, RBF and Polynomial. We have used RBF kernel for model generation. Before generating models from training set, a thorough search for two parameters "C" and gamma was carried out. The values for which maximum accuracy obtained after 10 folds cross validation are then used for model generation.