School of Physics, Beijing Institute of Technology, Beijing 100081, People's Republic of China.
Nordita, Stockholm University, Roslagstullsbacken 23, SE-106 91 Stockholm, Sweden.
J Chem Phys. 2019 Jun 14;150(22):225103. doi: 10.1063/1.5082627.
We inquire to what extent can the geometry of protein peptide plane and side chain atoms be reconstructed from the knowledge of Cα time evolution. Due to the lack of experimental data, we analyze all atom molecular dynamics trajectories from the Anton supercomputer, and for clarity, we limit our attention to the peptide plane O atoms and side chain Cβ atoms. We reconstruct their positions using four different approaches. Three of these are the publicly available reconstruction programs Pulchra, Remo, and Scwrl4. The fourth, Statistical Method, builds entirely on the statistical analysis of Protein Data Bank structures. All four methods place the O and Cβ atoms accurately along the Anton trajectories; the Statistical Method gives results that are closest to the Anton data. The results suggest that when a protein moves under physiological conditions, its all atom structures can be reconstructed with high accuracy from the knowledge of the Cα atom positions. This can help to better understand and improve all atom force fields, and advance reconstruction and refinement methods for reduced protein structures. The results provide impetus for the development of effective coarse grained force fields in terms of reduced coordinates.
我们想知道，从 Cα 时间演化的知识中，在多大程度上可以重建蛋白质肽平面和侧链原子的几何形状。由于缺乏实验数据，我们分析了 Anton 超级计算机上的所有原子分子动力学轨迹，为了清晰起见，我们将注意力限制在肽平面 O 原子和侧链 Cβ 原子上。我们使用四种不同的方法来重建它们的位置。其中三种是公开可用的重建程序 Pulchra、Remo 和 Scwrl4。第四种，统计方法，完全基于对蛋白质数据库结构的统计分析。所有四种方法都能准确地将 O 和 Cβ 原子放置在 Anton 轨迹上；统计方法给出的结果与 Anton 数据最接近。结果表明，当蛋白质在生理条件下运动时，可以从 Cα 原子位置的知识中以高精度重建其全原子结构。这有助于更好地理解和改进全原子力场，并推进简化蛋白质结构的重建和细化方法。该结果为基于简化坐标的有效粗粒力场的发展提供了动力。
