A similarity computing algorithm for proteins

Most of the existing algorithms for protein similarity comparison focus on the sequence comparison and structure comparison. However, the 3D structure of a protein is determined by the force field generated by all of its atoms in essence. Thus the similarity of force field implies the similarity of 3D structure. In this paper, we propose a novel approach to compare the similarity of protein´s force fields. First, we use blurred map to sample the force field into volumetric data sets. Second, the volume data set is resampled into a unified resolution. Third, the data set is band-pass filtered and quantized to reveal its physical attributes. The resulting voxels are then normalized into a canonical coordinate system concerning the center of mass and scale. Subsequently, a series of uniformly spaced concentric shells around the center of mass are constructed, based on which spherical harmonics analysis (SHA) is applied. The coefficient of SHA constitutes rotation invariant spectrum descriptors which are used to measure the similarity between two data sets. The algorithm has been performed on a set of proteins (taken from PDB) and the preliminary results are fairly inspiring.