Supervised Dimensionality Reduction Method for Predicting Membrane Proteins Types

With the rapid increase of protein sequences in the post-genomic age, the need for an automated and accurate tool to predict membrane protein types becomes increasingly important. Many efforts have been tried. Most of them aim to find the optimal classification scheme and less of them take the simplifying the complexity of biological system into consideration. This work shows how to decrease the complexity of biological system with the supervised DR (Dimensionality Reduction) method by transforming the original high-dimensional feature vectors into the low-dimensional feature vectors. Moreover, a powerful sequence encoding scheme by fusing PSSM (Position-Specific Score Matrix) and PseAA (Pseudo Amino Acid) method is used to represent the protein samples. Then, the K-NN (K-Nearest Neighbor) classifier is employed to identify the membrane protein types based on their reduced low-dimensional feature vectors. As a result, the jackknife and re-substitution test success rates on this model reach 85.2% and 92.6% respectively, and suggesting that the proposed approach is very promising for predicting membrane proteins types.