Prediction of protein folding types from amino acid composition by correlation angles.

A protein is usually classified into one of the following four folding types: all-α, all-β,α +β, andα/β. On the other hand, a protein can also be expressed as a vector of a 20-D space, in which its 20 components are defined by the composition of its 20 amino acids, respectively. Thus, the similarity between any two proteins in their amino acid composition can be quantitatively described in terms of their mutual projection in the 20-D space. The larger the projection value between the two vectors is, the higher their similar extent would be. Based on such a physical picture, a new method, the maximum projection method, has been proposed for predicting the folding type of a protein according to its amino acid composition. In comparison with the existing methods, the new method has the merits of yielding a higher rate of correct prediction, displaying a more intuitive physical picture, and being convenient in application.