Computer models of protein mutants for predictions of behaviors of protein folding and aggregation

点击翻译按钮获取中文摘要

In protein engineering protein's sequence structures are altered by substituting one or more residues on wild type proteins to produce mutants. The mutants often possess different native structures, and exibit different folding and binding behaviors than the original wild type proteins. Computer simulation models have been instrumental in understanding this medically important and scientifically revealing field. However, sophisticated software such as AMBER, CHARMM and GROMACS are too computationally intensive, while simplified Go models requires knowledge of the native structures, which are usually not available for mutants. This project constructs Go model of mutants based on the known (usually) structure of the wild type proteins. The novel method uses advance simulation software to construct the ground state mutant structures, and then employs computationally efficient Go models to study the folding and aggregation (binding) mechanisms of the mutants. If successful, the method will help to guide protein engineering experiments, and lead to designed proteins whose folding speed approach the so-called "protein folding speed limit", or to protein systems with altered functional roles.