计算生物学
生物信息学
突变体
计算机科学
突变
化学
生物
生物化学
基因
作者
Dibyendu Mondal,Vesselin Kolev,Arieh Warshel
标识
DOI:10.1021/acscatal.0c01206
摘要
Computer-aided enzyme design is a field of great potential importance for biotechnological applications, medical advances, and a fundamental understanding of enzyme action. However, reaching a predictive ability in this direction is extremely challenging. It requires both the ability to predict quantitatively the activation barriers in cases where the structure and sequence are known and the ability to predict the effect of different mutations. In this work, we propose a protocol for predicting reasonable starting structures of mutants of proteins with known structures and for calculating the activation barriers of the generated mutants. Our approach also allows us to use the predicted structures of the generated mutant to predict structures and activation barriers for subsequent set of mutations. This protocol is used to examine the reliability of the in silico directed evolution of Kemp eliminase and haloalkane dehalogenase. We also used the results of single and double mutations as a base for predicting the effect of transition-state stabilization by multiple concurrent mutations. This strategy seems to be useful in creating an activity funnel that provides a qualitative ranking of the catalytic power of different mutants.
科研通智能强力驱动
Strongly Powered by AbleSci AI