A method for matching sequences of protein secondary structures

Alignment of specific regions of two biological molecules is a basic method for determination how similar these two molecules are. There are several methods of optimal alignment that were developed through many years. However, they are dedicated for nucleotide sequences of DNA⁄RNA or amino acid sequences of proteins. Since the construction of proteins can also be analyzed at the level of secondary structure (and higher), we need a comparative method, which would allow us to determine the similarity between biological particles at this level and express it through the appropriate similarity measure. For this reason, we have modified an existing Smith–Waterman method towards matching sequences of secondary structures elements (SSEs). In the paper, we present our modification to the method. We also describe how we find several alternative and equally optimal alignment paths on the basis of the characteristics of compared sequences. Presented alignment method is used in the PSS–SQL language, which allows searching a database in order to find proteins having secondary structures similar to the structural pattern specified by a user.