Coarse-Grained Modelling of DNA-Protein Interactions

蛋白质在每一个活的有机体中执行大量的基本生物功能。在细胞中，它们不是作为单独的实体发挥功能，而是与其他生物大分子复合。因此，现代结构基因组学的重点正在迅速转向大分子DNA-蛋白质复合物的研究，这是许多感染性疾病，遗传性疾病和癌症的起源。虽然小角X射线散射（SAXS）等已建立的实验高分辨率技术提供了DNA-蛋白质复合物的静态视图，计算机模拟通常是能够提供这些复合物如何执行其功能的关键动态信息的唯一工具。粗粒度（CG）模型，再现相关的结构特征和热力学性质，大大降低了计算成本在这里发挥了特别重要的作用。由于生物学以具有自然界中最大和最复杂的结构而闻名，CG模型不仅是原子方法的有效替代方案。它们对于生物大分子如长DNA链和大蛋白质的毫秒级及更长时间的建模是不可或缺的。在过去的十年中，每个独立的领域都取得了惊人的进展，一方面是DNA的CG建模，另一方面是蛋白质的CG建模。oxDNA模型[1]（见方法论）已经成为DNA和RNA的主要CG模型。类似地，关联记忆、水介导、结构和能量模型（AWSEM，参见方法）是迄今为止最成功的CG蛋白模型之一。这两个模型都在大规模原子/分子大规模并行模拟器（LAMMPS，见方法）中实现，这是一个全球社区分子动力学代码，但不能结合使用。这使得DNA-蛋白质相互作用的新兴领域无法进入，这在原则上比DNA和蛋白质本身要广泛得多。本项目将开发oxDNA和AWSEM模型耦合的理论概念和软件实现。通过结合使用两个单独即将推出的模型和所产生的软件，它将构成DNA-蛋白质相互作用CG建模新兴领域的变革性一步，实现长时间和大长度尺度的模拟，并将结构生物学，DNA和蛋白质建模，合成生物学和纳米技术中不断增长的科学界聚集在一起。新功能将直接应用于与抗菌素耐药性和DNA纳米技术相关的项目。

Proteins perform a vast array of essential biological functions within every living organism. In the context of the cell, they do not perform their function as individual entities, but in complexes with other biomacromolecules. Consequently, the focus of modern structural genomics is rapidly shifting toward the study of macromolecular DNA-protein complexes, which are at the origin of many infectious diseases, genetic disorders, and cancer.While established experimental high-resolution techniques like small-angle X-ray scattering (SAXS) provide a static view of DNA-protein complexes, computer simulations are often the only tool capable of providing key dynamic information of how these complexes perform their functions. Coarse-grained (CG) models, which reproduce the relevant structural features and thermodynamic properties at greatly reduced computational costs play here a particularly important role. As biology is known for having the largest and most complex structures in nature, CG models are not only an efficient alternative to atomistic approaches. They are indispensable for modelling biomacromolecules such as long DNA strands and large proteins on timescales in the millisecond range and beyond.During the last decade striking progress has been made in each separate field, CG modelling of DNA on one side, and CG modelling of proteins on the other side. The oxDNA model [1] (see Methodology) has emerged as the leading CG model of DNA and RNA. Similarly, the Associative Memory, Water-Mediated, Structure and Energy Model (AWSEM, see Methodology) is among the most successful CG protein models to date. Both models are implemented in the Large-scale Atomic/Molecular Massively Parallel Simulator [2 - 4] (LAMMPS, see Methodology), a global community molecular dynamics code, but cannot be used in conjunction. This leaves the emerging field of DNA-protein interactions inaccessible, which is in principle much broader than DNA and proteins alone.This project will develop both the theoretical concepts and software implementations of a coupling of the oxDNA and AWSEM models. Through the combined use of the two individually forthcoming models and the produced software it will constitute a transformative step in the emerging field of CG modelling of DNA-protein interactions, enable simulations on very long time and large length scales, and bring together growing scientific communities in structural biology, DNA and protein modelling, synthetic biology and nanotechnology. The new functionality will be directly applied to projects related to antimicrobial resistance and DNA nanotechnology.