蛋白质动力学中若干涌现特征的研究

发布时间：2018-03-29 17:36

  本文选题：临界性　切入点：蛋白质　出处：《南京大学》2016年博士论文

【摘要】：在复杂系统中常常具有一些从系统整体层次展现出来的、无法还原为系统较低层次上简单因素的特征或规律,这些从系统整体层次表现出来的现象即为复杂系统的涌现行为.近年来,越来越多的实验和观测都表明,生命系统作为一种典型的非平衡系统,常常通过自组织临界工作在某种与相变的“临界点”类似的状态上.复杂系统在临界点附近可能涌现出许多重要的物理行为,尤其对生命系统而言,在临界点附近通常可以较好地保持“稳定性”和“适应性”之间的平衡.例如在各种生物的集体行为和各类复杂的生物网络中,系统既需要保持稳定的组织性,也需要对环境中的微小变化保持敏感性,而处在临界态的系统正好可以满足这两方面的条件：构成系统的个体之间仅仅存在短程的相互作用,就可以使得整个系统(例如动物群体的集体行为和神经动力学)却表现出长程的关联.蛋白质是生命活动的承载者,在生物体内发挥多种多样的功能.对于蛋白质分子而言,它们既具有稳定的天然态结构,又在行使其功能时足够灵活和敏感,在发生构象变化时可以保持一定的灵活性,使得分子可以发生大尺度的构象变化.这两种基本要求看起来相互矛盾,而当体系处在临界点附近时,其稳定性和灵活性有可能共存.我们对蛋白质的天然态结构进行了研究.我们通过对核磁共振方法测定的蛋白质结构出发,从这些数据中提取出蛋白质在天然态结构附近的涨落特征,通过计算这些涨落中的关联,直接证明了蛋白质天然态涨落的临界特征.在蛋白质的天然态涨落中,涨落的关联长度与蛋白质分子的尺寸成正比,而这些天然态结构中存在着大量的幂律.我们还通过统计说明这些临界的特征对于不同结构、不同功能的蛋白质分子是普遍存在的.蛋白质天然态附近的临界动力学与蛋白质的进化有着紧密的联系：具有临界特征的蛋白质在进化的过程中具有适应度的优势,它们既具有进化上和结构上的稳定性,又具有可进化性以及发挥生物学功能的能力.为了进一步分析这种临界性的起源,基于蛋白质的序列进化和天然态结构的信息,采用网络分析的方法,深入研究了共进化直接耦合网络和残基接触网络(弹性网络模型)的性质.目前的研究表明这两种不同类型的网络本身是高度相似的.我们证明,对这些网络而言,在拓扑空间中,蛋白质分子的无标度涨落特征仍然是存在的.此外,通过以无序堆积的粒子形成的网络结构作为对照,我们证明了蛋白质的这种临界性与其天然态结构是相关的.我们还发现：氨基酸残基的度度关联表明这一网络具有无标度网络的特征,并且残基接触网络表现出模块化和层级化的特征,而跟随机堆积相比,氨基酸残基的堆积表现出高聚集系数和非小世界的特点,反映到分子的振动模式上,则对应于相比随机堆积更低的最低频模式,这些低频模式的频率与蛋白质的形状和表面积等是直接相关的.我们的这一分析揭示了蛋白质天然态结构临界性的序列进化和结构基础.蛋白质天然态涨落表现出明显的临界特征,因而分子内的关联长度会随着分子尺寸的增加而增加,但分子内的相互作用力程并不会发生改变.为了深入研究这一现象的原因,我们对蛋白质分子内的这种长程关联进行了更深入的统计分析.我们利用了关联矩阵和相互作用矩阵的互逆关系,设计了一种自洽的方法,对残基间距离依赖的相互作用进行了推断,这一方法利用了残基本身相互作用的一些信息,计算的效率比极大熵方法等更高.我们的这一方法可以得到残基间相互作用的许多重要特征,且其推断的相互作用强度与统计势也是相关的.我们关于蛋白质动力学中若干涌现特征的研究揭示了蛋白质结构和功能之间的基本关系.这一结果表明：蛋白质的天然态结构不只要求其处在能量面的最小值处,并且,蛋白质的天然态结构还保持着其动态的柔性.在自然选择中被选择的这些蛋白质的天然态结构可以保证氨基酸残基间长程关联的出现,这对于蛋白质发生构象变化和保持稳定的天然态结构都是很重要的.我们的研究结果还对蛋白质设计、蛋白质的聚集以及蛋白质的折叠和进化等诸多问题具有重要的参考价值.
[Abstract]:In a complex system often has some from the system level to show out, can not be reduced to a relatively low level of simple system features or rules of these factors, from the system level performance of the whole phenomenon is the emergence of behavior of complex systems. In recent years, more and more experiments and observations have shown that as a life system non equilibrium system of typical, often through the self organized critical work in a phase transition and "critical point" of similar state. The complex system may be the emergence of many important physical behavior near the critical point, especially on the life of the system, in the vicinity of the critical point is usually between good stability and adaptability "the balance. For example in a variety of biological collective behavior and all kinds of complex biological networks, the system needs to maintain the stability of the organization, but also the need for small change in the environment Keep sensitive in the critical state of the system can satisfy the two conditions: only the existence of interaction between the short-range system which consists of the individual, can make the whole system (such as animal collective behavior and neural dynamics) has shown long-range correlation. The protein is the bearer of life activities, play a variety of functions in organism. For protein molecules, they have not only the natural state of stable structure, and in the exercise of its function is sufficiently flexible and sensitive, can maintain a certain flexibility in the conformational change, the molecular conformational changes can occur in large scale. The two basic requirements seem contradictory, but when the system is near the critical point, its stability and flexibility. We may coexist on the protein native structure is studied. We based on NMR Protein structure determination method of resonance, extract the fluctuation characteristics of protein in the vicinity of the natural state of the structure from the data, by calculating the correlation of these fluctuations in the direct proof of the critical characteristics of the natural state of protein fluctuations. In the natural state of fluctuation of protein, fluctuation correlation length and protein size is proportional to the there are a large number of these natural power-law structure. We also illustrate these critical features through statistics for different structures, different functions of the protein molecule is universal. The natural state of critical dynamics near the protein and protein evolution are closely linked with the critical feature protein with fitness advantage in evolution in the process, they both stability and structure evolution, and evolution also has the ability to play the biological functions. Further analysis of the origins of this critical, protein sequence evolution and natural state structure based on information, using the method of network analysis, in-depth study of the direct coupling network co evolution and residue contact network (elastic network model) properties. The present study shows that these two different types of network itself is highly similar. We show that, for these networks, in the topology space, protein scale-free fluctuation characteristics still exists. In addition, the network structure formed by the disordered accumulation of particles as the control, we permit this critical and natural state is related to protein structure and clear. We also found that the degree of association amino acid residues show that this network has the characteristics of scale-free network, and the residue contact network shows characteristics of modular and hierarchical, with random accumulation than amino acid The accumulation of residues showed high accumulation coefficient and non small world, to reflect the molecular vibrational modes, corresponding to the low frequency mode compared to random packing of lower frequency and low frequency mode, the protein shape and surface area are directly related. This analysis we revealed the sequence evolution and structure based protein native structure of protein. The critical natural state fluctuation showed critical characteristics, and the correlation length in the molecule will increase with the increase of molecular size, but the interaction range in the molecule and does not change. In order to study the reasons of this phenomenon, we on protein in the the long-range correlation has been statistically analyzed deeply. We use the inverse relationship between the incidence matrix and the interaction matrix, design a self consistent method, the distance between the residues in The interaction of Lai was deduced, this method uses some information interaction residues, the calculation efficiency is higher than the maximum entropy method. This method we can get many important features of the interaction between residues, interaction strength and potential and the statistical inference is also relevant. Us about some emerging features of protein dynamics research reveals the basic relationship between protein structure and function. The results show that the natural state of the protein structure not only requires its energy in the minimum value, and the natural state of structural protein keeps its dynamic flexibility. These proteins are natural state structure selection in natural selection can guarantee the emergence of long-range correlations among amino acid residues, the protein conformational change and maintain the stability of the native state structure is very important Our results also have important reference value for protein design, protein aggregation, protein folding and evolution, and many other problems.

【学位授予单位】：南京大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：Q615
，

本文编号：1682140