Computer simulations of lipids

我的长期目标是发展精确的计算模型来描述膜组分之间的分子相互作用，并利用这些模型来解决脂质双层和单层的生物物理学和生物化学领域的基本问题。在过去的五年中，脂类的计算机模拟和更普遍的膜模型有了显著的发展，现在已经成为一种广泛使用的技术来研究脂类的物理化学和生物物理学，无论是本身还是与实验相结合。自20世纪90年代该领域成立以来，我的团队一直在为其做出贡献，当时的计算机能力比今天低5个数量级，模型也远没有今天复杂。在过去的五年中，我们在改进脂质模型，增加模型的复杂性以及向生物系统的复杂性迈进方面取得了重大进展，同时继续我们在控制良好的模型系统上的工作，包括简单的脂质混合物，胶束和单束。我们主要使用分子动力学模拟和自由能计算，基于脂质和其他分子的原子和粗粒度描述，辅以其他方法，如电子结构计算和连续介质静电计算。我们共同开发了Martini，这是目前最流行的粗粒度分子模拟模型。在我们的研究计划中，我们将追求两个主要方向。第一个目标是扩展我们的计算工具包来模拟复杂的膜，包括进一步开发Martini，使用结合分子动力学/蒙特卡罗方法更有效地混合脂质混合物的计算方法，以及在膜模拟的可视化和分析中使用虚拟现实软件。第二个方向是由关于膜结构和生物膜的许多组成部分的作用的基本问题驱动的。我们将研究作为组成函数的脂质混合物的复杂特性，包括模型膜，如生物膜，在形成其基本双层结构的两个小叶中的每一个都有不同的组成。与这个问题相关的是，大多数细胞膜的主要成分胆固醇如何在膜内重新分布并与其他成分相互作用，以及脂质分布和蛋白质构象如何通过膜曲率相互耦合。这两个方向是相互关联的，因为改进的计算工具对于回答基本问题和更紧密地联系实验和模拟是必不可少的。用于研究基本问题的系统通常也可用于开发、测试和验证模型和新方法。该计划将改进计算工具，可用于广泛的未来研究，以及洞察生物膜的基本组成部分。

My long-term aim is the development of accurate computational models to describe the molecular interactions between membrane components and the use of these models to solve fundamental problems in the area of the biophysics and biochemistry of lipid bilayers and monolayers. Computer simulations of lipids and more generally membrane models have grown significantly in the past five years and are now a widely used technique to study the physical chemistry and biophysics of lipids, both by themselves and in combination with experiment. My group has contributed to this area since its inception in the 1990s, when computer power was ca. 5 orders of magnitude less than today and the models much less sophisticated. In the past five years we have made significant progress in improving lipid models, increasing the complexity of our models, and moving closer towards the complexity of biological systems while continuing our work on well-controlled model systems including simple lipid mixtures, micelles and bicelles. We use primarily molecular dynamics simulations and free energy calculations, based on atomistic and coarse-grained descriptions of lipids and other molecules, supplemented by other methods such as electronic structure calculations and continuum electrostatics calculations where required. We co-develop Martini, currently the most popular coarse-grained model for molecular simulation. We will pursue two major directions in our research program. The first is aimed at expanding our computational toolkit to model complex membranes and includes the further development of Martini, computational methods to more efficiently sample mixing of lipid mixtures using a combined molecular dynamics/Monte Carlo method, and the use of virtual reality software in visualization and analysis of membrane simulations. The second direction is driven by fundamental questions about membrane structure and the role of the many components of biological membranes. We will investigate the complex properties of lipid mixtures as a function of composition, including model membranes that, like biological membranes, have a different composition in each of the two leaflets that form their basic bilayer structure. Related to this problem is how cholesterol, a major component of most cell membranes, redistributes within the membrane and interacts with other components, and how lipid distributions and protein conformation are coupled to each other through membrane curvature. These two directions are interrelated as improved computational tools are essential to answer the fundamental questions and to more closely link experiment and simulations. The systems used to study fundamental questions typically are also useful to develop, test, and validate models and new methods. The program will result in improved computational tools that can be used in a broad range of future studies, as well as insight in the fundamental building blocks of biological membranes.