High performance computing and communications grand challenges program

Technical Report · Sat Oct 01 04:00:00 EDT 1994

DOI:https://doi.org/10.2172/378965· OSTI ID:378965

Solomon, J E; Barr, A; Chandy, K M; Goddard, III, W A; Kesselman, C

The so-called protein folding problem has numerous aspects, however it is principally concerned with the {ital de novo} prediction of three-dimensional (3D) structure from the protein primary amino acid sequence, and with the kinetics of the protein folding process. Our current project focuses on the 3D structure prediction problem which has proved to be an elusive goal of molecular biology and biochemistry. The number of local energy minima is exponential in the number of amino acids in the protein. All current methods of 3D structure prediction attempt to alleviate this problem by imposing various constraints that effectively limit the volume of conformational space which must be searched. Our Grand Challenge project consists of two elements: (1) a hierarchical methodology for 3D protein structure prediction; and (2) development of a parallel computing environment, the Protein Folding Workbench, for carrying out a variety of protein structure prediction/modeling computations. During the first three years of this project, we are focusing on the use of two proteins selected from the Brookhaven Protein Data Base (PDB) of known structure to provide validation of our prediction algorithms and their software implementation, both serial and parallel. Both proteins, protein L from {ital peptostreptococcus magnus}, and {ital streptococcal} protein G, are known to bind to IgG, and both have an {alpha} {plus} {beta} sandwich conformation. Although both proteins bind to IgG, they do so at different sites on the immunoglobin and it is of considerable biological interest to understand structurally why this is so. 12 refs., 1 fig.

• Research Organization: California Inst. of Tech., Pasadena, CA (United States)
• Sponsoring Organization: USDOE, Washington, DC (United States)
• DOE Contract Number: FG03-92ER25134
• OSTI ID: 378965
• Report Number(s): DOE/ER/25134--T2; ON: DE96013014
• Country of Publication: United States
• Language: English

Related Subjects

55 BIOLOGY AND MEDICINE
BASIC STUDIES
99 GENERAL AND MISCELLANEOUS
COMPUTER CALCULATIONS
MOLECULAR BIOLOGY
MONTE CARLO METHOD
PARALLEL PROCESSING
PROTEIN STRUCTURE