Sayyed-Ahmad Abdallah, Tuncay Kagan, Ortoleva Peter J
Center for Cell and Virus Theory, Department of Chemistry, Indiana University, Bloomington, Indiana 47405, USA.
J Comput Chem. 2004 Jun;25(8):1068-74. doi: 10.1002/jcc.20039.
The Poisson-Boltzmann (PB) equation has been extensively used to analyze the energetics and structure of proteins and other significant biomolecules immersed in electrolyte media. A new highly efficient approach for solving PB-type equations that allows for the modeling of many-atoms structures such as encountered in cell biology, virology, and nanotechnology is presented. We accomplish these efficiencies by reformulating the elliptic PB equation as the long-time solution of an advection-diffusion equation. An efficient modified, memory optimized, alternating direction implicit scheme is used to integrate the reformulated PB equation. Our approach is demonstrated on protein composites (a polio virus capsid protomer and a pentamer). The approach has great potential for the analysis of supramillion atoms immersed in a host electrolyte.
泊松-玻尔兹曼（PB）方程已被广泛用于分析浸没在电解质介质中的蛋白质及其他重要生物分子的能量学和结构。本文提出了一种求解PB型方程的高效新方法，该方法能够对细胞生物学、病毒学和纳米技术中遇到的多原子结构进行建模。我们通过将椭圆型PB方程重新表述为平流扩散方程的长时间解来实现这些效率提升。采用一种高效的改进型、内存优化的交替方向隐式格式对重新表述的PB方程进行积分。我们的方法在蛋白质复合物（脊髓灰质炎病毒衣壳原体和五聚体）上得到了验证。该方法在分析浸没在主体电解质中的超百万原子方面具有巨大潜力。
