Department of Physics, Department of Biochemistry, and MU Informatics Institute , University of Missouri , Columbia , Missouri 65211 , United States.
J Chem Theory Comput. 2018 Apr 10;14(4):2230-2239. doi: 10.1021/acs.jctc.7b01228. Epub 2018 Mar 9.
Coarse-grained RNA folding models promise great potential for RNA structure prediction. A key component in a coarse-grained folding model is the force field. One of the challenges in the coarse-grained force field calculation is how to treat the correlation between the different degrees of freedoms. Here, we describe a new approach (IsRNA) to extract the correlated energy functions from the known structures. Through iterative molecular dynamics simulations, we build the correlation effects into the reference states, from which we extract the energy functions. The validity of IsRNA is supported by the close agreement between the simulated Boltzmann-like probability distributions for all the structure parameters and those observed from the experimentally determined structures. The correlated energy functions derived here may provide a new tool for RNA 3D structure prediction.
粗粒化 RNA 折叠模型有望在 RNA 结构预测方面取得重大进展。在粗粒化折叠模型中,力场是一个关键组成部分。在粗粒化力场计算中面临的挑战之一是如何处理不同自由度之间的相关性。在这里,我们描述了一种从已知结构中提取相关能量函数的新方法(IsRNA)。通过迭代分子动力学模拟,我们将相关效应构建到参考状态中,从中提取能量函数。通过模拟的玻尔兹曼似然概率分布与从实验确定的结构中观察到的所有结构参数之间的紧密一致性,证明了 IsRNA 的有效性。这里得到的相关能量函数可能为 RNA 三维结构预测提供新的工具。