Challenges in constructing accurate methods for hydrogen transfer reactions in large biological assemblies: rare events sampling for mechanistic discovery and tensor networks for quantum nuclear effects.,Faraday Discussions

当前位置： X-MOL 学术 › Faraday Discuss. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Challenges in constructing accurate methods for hydrogen transfer reactions in large biological assemblies: rare events sampling for mechanistic discovery and tensor networks for quantum nuclear effects.
Faraday Discussions ( IF 3.3 ) Pub Date : 2019-12-16 , DOI: 10.1039/c9fd00071b
Nicole DeGregorio ₁ , Srinivasan S Iyengar

Affiliation

We present two methods that address the computational complexities arising in hydrogen transfer reactions in enzyme active sites. To address the challenge of reactive rare events, we begin with an ab initio molecular dynamics adaptation of the Caldeira-Leggett system-bath Hamiltonian and apply this approach to the study of the hydrogen transfer rate-determining step in soybean lipoxygenase-1. Through direct application of this method to compute an ensemble of classical trajectories, we discuss the critical role of isoleucine-839 in modulating the primary hydrogen transfer event in SLO-1. Notably, the formation of the hydrogen bond between isoleucine-839 and the acceptor-OH group regulates the electronegativity of the donor and acceptor groups to affect the hydrogen transfer process. Curtailing the formation of this hydrogen bond adversely affects the probability of hydrogen transfer. The second part of this paper deals with complementing the rare event sampled reaction pathways obtained from the aforementioned development through quantum nuclear wavepacket dynamics. Essentially the idea is to construct quantum nuclear dynamics on the potential surfaces obtained along the biased trajectories created as noted above. Here, while we are able to obtain critical insights on the quantum nuclear effects from wavepacket dynamics, we primarily engage in providing an improved computational approach for efficient representation of quantum dynamics data such as potential surfaces and transmission probabilities using tensor networks. We find that utilizing tensor networks yields an accurate and efficient description of time-dependent wavepackets, reduced dimensional nuclear eigenstates and associated potential energy surfaces at much reduced cost.

中文翻译：

构建大型生物组件中氢转移反应的准确方法所面临的挑战：用于机制发现的稀有事件采样和用于量子核效应的张量网络。

我们提出了两种解决酶活性位点氢转移反应中产生的计算复杂性的方法。为了解决反应性稀有事件的挑战，我们从对Caldeira-Leggett系统-浴哈密顿量的从头算分子动力学适应开始，并将这种方法应用于研究大豆脂氧合酶-1中氢转移速率确定步骤的研究。通过直接应用此方法来计算经典轨迹的合奏，我们讨论了异亮氨酸839在调节SLO-1中主要氢转移事件中的关键作用。值得注意的是，异亮氨酸-839和受体-OH基团之间氢键的形成调节了供体和受体基团的电负性以影响氢转移过程。减少该氢键的形成不利地影响了氢转移的可能性。本文的第二部分涉及通过量子核波包动力学对上述发展过程中获得的稀有事件采样反应路径进行补充。本质上，该想法是在沿如上所述产生的有偏向轨迹获得的潜在表面上构建量子核动力学。在这里，尽管我们能够从波包动力学中获得关于量子核效应的批判性见解，但我们主要致力于提供一种改进的计算方法，以有效地表示使用张量网络的量子动力学数据（例如，潜在的表面和传输概率）。

更新日期：2019-12-17

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11