The hierarchy in the natural protein sequences and it's applications

A method was proposed for the analysis of protein sequences. The method allows identifying the hierarchical organization of structural information of protein sequences. Research of the entropy characteristics of natural polypeptide sequences [1] reveals maximal size of protein sequence with minimal and constant entropy value as five adjacent amino acids. Such a group of five neighboring residues was used as a minimal unit of protein sequence and has been called the “information unit”. Based on information units representation, special mathematical approach for protein sequence analysis was called “Analysis of Information Structure” method (ANIS method) [2]. ANIS method consists of several steps. Initially, the frequency of occurrence in the database of non-homologous protein sequences is determine for each of the information units that comprise the studied protein sequence. Then, “population profile” is built based on the frequency of occurrence of information units. On the last step “population profile” is decomposed into positive defined symmetric functions with different half-width. Hierarchically organized information structure of the protein sequence is a result of the ANIS method. The ANIS method enabling to reveal the hierarchical organization of structural information contained in amino acid sequence of proteins. The approach was tested in the set of experimental studies by protein engineering. For example, functionally important fragments of heat shock protein (hHSP70), human tumor necrosis factor (hTNF) and protein gp181 from phage KZ were obtained. The proposed approach can be used for de novo protein design. Applications of ANIS method in protein engineering [3, 4], studies of structural organization of enzymes [5, 6] and protein-protein complexes [7] are described. In general method described in [8, 9].