Protein secondary structure prediction based on ramachandran maps

Abstract

High sequence similarity of two proteins does not imply the relation in their structural similarity. In this paper, we propose a protein secondary structure prediction method based on Ramachandran maps. According to the distribution of Ramachandran plot on φ and ψ backbone conformational angles, the series of backbone dihedral angles are regarded as a sequence of input states. Individual residues are classified into different secondary structures. In one of our results, protein hemoglobin 1A6M aligns 2FAM sequence sharing 11.74% similarity. However, 1A6M shares 81.82% of the common structure with 2FAM. Results display that our proposed method provides a promotive advantage on protein secondary structure prediction, without depending on sequence similarity. © 2008 Springer-Verlag Berlin Heidelberg.