Tabatabaei Ghomi Hamed, Topp Elizabeth M, Lill Markus A
Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, IN, USA.
Department of Industrial and Physical Pharmacy, Purdue University, West Lafayette, IN, USA.
J Mol Model. 2016 Sep;22(9):206. doi: 10.1007/s00894-016-3066-1. Epub 2016 Aug 8.
Amyloid fibrils are important in diseases such as Alzheimer's disease and Parkinson's disease, and are also a common instability in peptide and protein drug products. Despite their importance, experimental structures of amyloid fibrils in atomistic detail are rare. To address this limitation, we have developed a novel, rapid computational method to predict amyloid fibril structures (Fibpredictor). The method combines β-sheet model building, β-sheet replication, and symmetry operations with side-chain prediction and statistical scoring functions. When applied to nine amyloid fibrils with experimentally determined structures, the method predicted the correct structures of amyloid fibrils and enriched those among the top-ranked structures. These models can be used as the initial heuristic structures for more complicated computational studies. Fibpredictor is available at http://nanohub.org/resources/fibpredictor .
淀粉样纤维在阿尔茨海默病和帕金森病等疾病中具有重要意义，并且也是肽和蛋白质药物产品中常见的不稳定性因素。尽管它们很重要，但原子水平详细的淀粉样纤维实验结构却很少见。为了克服这一局限性，我们开发了一种新颖、快速的计算方法来预测淀粉样纤维结构（Fibpredictor）。该方法将β-折叠模型构建、β-折叠复制和对称操作与侧链预测及统计评分函数相结合。当应用于九个具有实验确定结构的淀粉样纤维时，该方法预测出了正确的淀粉样纤维结构，并在排名靠前的结构中富集了这些结构。这些模型可作为更复杂计算研究的初始启发式结构。Fibpredictor可在http://nanohub.org/resources/fibpredictor获取。
