Steinbrecher Thomas, Abel Robert, Clark Anthony, Friesner Richard
Schrödinger GmbH, Dynamostr. 13, 68165 Mannheim, Germany.
Schrödinger Inc., 120 West 45th Street, 17th Floor, New York, NY 10036, USA.
J Mol Biol. 2017 Apr 7;429(7):923-929. doi: 10.1016/j.jmb.2017.03.002. Epub 2017 Mar 6.
Protein side-chain mutation is fundamental both to natural evolutionary processes and to the engineering of protein therapeutics, which constitute an increasing fraction of important medications. Molecular simulation enables the prediction of the effects of mutation on properties such as binding affinity, secondary and tertiary structure, conformational dynamics, and thermal stability. A number of widely differing approaches have been applied to these predictions, including sequence-based algorithms, knowledge-based potential functions, and all-atom molecular mechanics calculations. Free energy perturbation theory, employing all-atom and explicit-solvent molecular dynamics simulations, is a rigorous physics-based approach for calculating thermodynamic effects of, for example, protein side-chain mutations. Over the past several years, we have initiated an investigation of the ability of our most recent free energy perturbation methodology to model the thermodynamics of protein mutation for two specific problems: protein-protein binding affinities and protein thermal stability. We highlight recent advances in the field and outline current and future challenges.
蛋白质侧链突变对于自然进化过程以及蛋白质治疗药物的工程设计都至关重要，而蛋白质治疗药物在重要药物中所占比例日益增加。分子模拟能够预测突变对诸如结合亲和力、二级和三级结构、构象动力学以及热稳定性等性质的影响。许多截然不同的方法已被应用于这些预测，包括基于序列的算法、基于知识的势函数以及全原子分子力学计算。采用全原子和显式溶剂分子动力学模拟的自由能微扰理论，是一种基于物理学的严谨方法，用于计算例如蛋白质侧链突变的热力学效应。在过去几年中，我们已着手研究我们最新的自由能微扰方法对两个特定问题进行蛋白质突变热力学建模的能力：蛋白质 - 蛋白质结合亲和力和蛋白质热稳定性。我们重点介绍了该领域的最新进展，并概述了当前和未来的挑战。
