STRUCTURE PREDICTION OF PEPTIDES VIA GLOBAL OPTIMIZATION

描述：这个项目的总体目标是改善我们的 对三维结构预测的认识 寡肽、多肽和蛋白质。根据他最近 (A)基于分解的全局优化方法， (B)凸函数变换的差分枝有界 全局优化方法，alphaBB，(C)GOP在算法中的应用 Lennard-Jones团簇结构预测，(D)Lennard-Jones团簇的应用 αBB到无环小分子，(E)αBB的应用 再加上ECEPP/3使天然产生的氨基酸和小分子 寡肽，(F)αBB与ECEPP/3偶联的应用 甲硫氨酸脑啡肽及其与模拟退火法和其他搜索方法的比较 方法，(G)α-BB在十甘氨酸中的应用，以及(H) 他在所考虑的系统上观察到的计算复杂性， 拟议的研究工作旨在确定(I)全球 多肽和蛋白质的最小势能构象，(Ii)低 多肽的能量构象接近全球最小构象，以及 (Iii)肽的势能超曲面的鞍点 接近全球最小多肽结构。弗洛达斯博士 计划重点实现以下目标： (A)研究中描述的确定性全局优化方法 关于(I)亮氨酸等寡肽的初步研究第3.3节。 脑啡肽，格拉米菌素S和梅利汀，(Ii)合并 溶剂化效应及其在甲硫氨酸五肽中的应用 脑啡肽和亮氨酸脑啡肽在水中的存在，(Iii) 相对熵和相对自由能的计算，以及(Iv) 受约束的多肽系统。(B)调查修改《 (A)中的全局优化方法，以便扩展到处理 多肽分子，如禽类胰腺多肽，模型 纤维蛋白，如胶原蛋白和球状蛋白，如 牛胰蛋白酶抑制剂。(C)研究全境的地形 发展和应用一种计算势能面的新方法 多肽能量超表面的鞍点测定 接近全球最低水平。(四)发展分布式计算 适用于(A)、(B)和(C)的全局优化方法的算法 将它们转化为多肽、多肽、蛋白质，并开发用于 预测多肽的三维结构。

DESCRIPTION: The overall aim of this project is to improve our understanding of the three dimensional structure prediction of oligopeptides, polypeptides, and proteins. Based upon his recent advances in: (a) decomposition-based global optimization approach, GOP, (b) branch and bound with difference of convex functions transformation global optimization method, alphaBB, (c) the application of GOP to the structure prediction of Lennard-Jones Clusters, (d) the application of alphaBB to small acyclic molecules, (e) the application of the alphaBB coupled with ECEPP/3 to the naturally occurring amino acids and small oligopeptides, (f) the application of alphaBB coupled with ECEPP/3 to Met-enkephalin and its comparison to simulated annealing and other search methods, (g) the application of alphaBB to Decaglycine, and (h) the computational complexity that he has observed on the considered systems, the proposed research work is directed at determining (i) the global minimum potential energy conformation of peptides and proteins, (ii) low energy conformations of peptides close to the global minimum one, and (iii) saddle points of the potential energy hypersurface of the peptides that are close to the global minimum peptide structure. Dr. Floudas plans to focus on the following objectives: (a) Study the deterministic global optimization approach described in the preliminary studies section 3.3 for (i) oligopeptides such as Leu- enkephalin, Gramicidin S, and Mellitin, (ii) the incorporation of solvation effects and the application to the pentapeptides of Met- enkephalin and Leu-enkephalin in the presence of water, (iii) calculations of the relative entropy and relative free energy, and (iv) constrained peptide systems. (b) Investigate modifications of the global optimization method in (a) so as to be extended to handle polypeptide molecules such as the avian pancreatic polypeptide, models of fibrous proteins such as collagen,a nd globular proteins such as bovine trypsin inhibitor. (c) Study the topography of the total potential energy surface by developing and applying a new method for the determination of saddle points of the peptide energy hyersurface that are close to the global minimum one. (d) Develop distributed computing algorithms for global optimization methods for (a), (b), and (c), apply them to olgopeptides, polypeptides, proteins, and develop tools for the prediction of three-dimensional structures of peptides.