Identification of GPI-(like)-Anchored Proteins by Using SVM

A new simple method for identification of GPI-(like)-anchored proteins at sequence level is proposed in this paper. As a binary classifier of GPI-(like)-anchored proteins and non GPI-(like)-anchored proteins, a supervised machine learning algorithm, support vector machine (SVM) and simple representation of C-terminus of protein primary sequences with mean hydrophobicity were used. Not merely does the classifier show high accuracy of 96.00% under 5-fold cross validation test, but also the A UC as a good summary of the performance of the classifier reaches to 0.97. In virtue of being based on SVM, computational efficiency and remarkable generalization ability of our classifier would be helpful for protein annotation in whole genomic-wide