Pattern recognition metric for comparison of protein structures based on amino acid sequences

An increase in the number of reported protein primary sequences has increased the need to extract structural and functional information from them. This growing need has precipitated the emergence of computational methods for predicting the structure and function of a protein on the basis of its amino acid sequence. This chapter describes basic pattern recognition metric for direct comparison of protein structural patterns solely on the basis of amino acid sequences. The core of the comparison metric is easily automated because it does not require many generalized rules or subjective parameters, which might need to be optimally adjusted; instead, the metric requires only knowledge of the amino acid sequence of a protein. To overcome the limitations of predictive methods that require smoothing and the defining of structures, the comparison metric quantifies comparisons among homologous and heterologous residues between two sequences with a nonheuristic ranking approach. Because the assignment of gap penalties, like operating windows, requires a subjective judgment, gaps are not inserted to optimize the alignment of sequences before comparing them.