Prediction of O-linked glycosylation sites in protein by independent component analysis

Glycosylation is one of the most important post-translation modifications steps in eukaryotic cell. In this paper, we propose a new approach based on independent component analysis (ICA) for prediction O-linked glycosylation site and pattern analysis. Principal component analysis (PCA) is first used to find significant uncorrelated components, and then ICA is used to extract independent components to construct a subspace (main basis) of protein sequence. The prediction is viewed as a 2-classes classification problem. The test protein vector is projected to each subspace. The protein sequence is classified into the nearest class by calculating the distance between the test vector and its projection on the subspace. The prediction accuracy of our proposed new approach is higher than that of other subspace methods based on PCA.