Prediction of protein subcellular location using the information entropy and the auto covariance transformation

The information of subcellular location is important to understand the functions of the proteins.Considerable efforts have been made for the precise prediction of protein subcellular location. However, the feature representation of protein sequences, a fundamental step in most of existing computational methods, is still a challenging task. In this paper, a new feature extraction method is proposed based on the information entropy and the auto covariance transformation. With information entropy, the distribution of each n-length amino acid sequence is depicted according to its positions in the input protein. Meanwhile, auto covariance transformation is applied to the position specific score matrix to measure the correlation between amino acid residues during the evolution process. Furthermore, the two descriptors described above are combined to improve the prediction performance of protein subcellular locations. The experimental results on three benchmark datasets show that the representation capability of the features is more powerful and the prediction is more accurate by applying our method.