Nonlinear Signal Analysis Approach to Protein Structure/Dynamics

Rush University Medical Center, Chicago IL

Investigators

Abstract

0240230 Zbilut This project takes a unique and somewhat novel approach by treating the amino acid code as a "signal." In so doing it uses recurrence analysis, a method derived from nonlinear dynamics. An important feature of this method is the absence of mathematical assumptions: proteins are relatively short, nonlinear and nonstationary, thus precluding the uninformed use of more traditional signal analysis methods. This strategy seeks to uncover unforeseen "singularities" of hydrophobicity in proteins and related function. The main steps of this signal analysis approach can be summarized as follows: 1) use of an hydrophobic code for primary structures; and 2) treatment of the hydrophobicity distribution along the sequence like a time series, with the corresponding use of nonlinear signal analysis techniques, in order to underpin fine position-dependent properties of the hydrophobicity profiles. It is posited that singularities in the code identify important loci at which proteins may fold correctly vs. aggregate. Moreover, the direction chosen is dependent upon local factors, or boundary conditions, which force a stochastic understanding of the process in terms of combinatorial probabilities. There is also a suggestion that these folding choices describe natural processes of aging and/or pathology The main aims of the project are: 1)the development of a preliminary taxonomy of these hydrophobic singularities; 2)a cataloguing of the singularities' protein context; i.e., a determination of their boundary conditions(e.g., secondary structure, absolute values, scaling etc.); and 3)an evaluation of their correlation with physico-chemical properties especially electrostatic variables The entire structure of the protein is fitted to perform many tasks but possibly only local patches (and consequently peculiar local hydrophobicity singularities) may be crucial for a given action or structure. From a chemical viewpoint, proteins, which are important elements in all forms of life, are made of amino acids. The majority of proteins "fold" as relatively small self-contained structures. Each protein found in nature has a specific three-dimensional structure and this structure is determined by the sequence of amino acids and their attendant degrees of aversion for water. This makes the particular linear arrangement of amino acids constituting a protein an efficient recipe for the solution of a chemico-physical problem that basically is the folding to a given unique three-dimensional structure in water solution. What the sequence must attain in order to be a real protein is basically to be water-soluble, having a well-defined (if not necessarily static) three-dimensional structure allowing for motions in solution (proteins do their work in a dynamical way by coordinated motions of their scaffolds) while at the same time maintaining their global shape. This is not an easy task and only a relative minority of linear amino acid arrangements are effective solutions to this problem. This is equivalent to saying that the "code" linking a sequence to its particular structure is not immediately apparent. Nevertheless it is known that the three-dimensional structure of a protein is in some (still obscure) way encoded in its amino acid order. Indeed, this question is more intriguing given the evidence that any protein can "misfold," causing possibly diseases such as Alzheimer's, "mad cow" or "aging" in general. Thus, the most basic problem in the sequence/structure puzzle is: "What particular linear arrangement of amino acids makes a real protein. This project seeks to answer this question by using unique "signal analysis" techniques which may uncover peculiarities in the sequence. Clearly, should distinctive conditions be discovered which promote particular protein folding behavior, enhanced understanding (and control) of pathology, senescence, and therapeutics could result. This grant is made under the Joint DMS/NIGMS Initiative to Support Research Grants in the Area of Mathematical Biology. This is a joint competition sponsored by the Division of Mathematical Sciences (DMS) at the National Science Foundation and the National Institute of General Medical Sciences (NIGMS) at the National Institutes of Health.