In this study, a novel sequence encoding scheme is introduced by fusing PseAA and PSSM. However, this sequence encoding scheme would correspond to a very high dimensional feature vector. A dimensionality reduction algorithm, the so-called NPE (Neighborhood Preserving Embedding) is introduced to extract the key features from the high-dimensional space. Finally, the K-NN (K-Nearest Neighbor) classifier is employed to identify the types of protein structures. Our jackknife test results thus obtained are quite encouraging, which indicate that the above methods are used effectively to deal with this complicated problem of predicting protein structural classes.