肺表面活性剂是一种吸附在肺泡表面的生物活性物质，主要由脂质和蛋白质组成。它由肺泡II型细胞组装并分泌到肺泡表面上，形成一层液体薄膜。肺表面活性剂膜是防御病原体或颗粒物吸入人体的第一道屏障，并能够降低肺泡表面张力，防止肺泡在呼吸过程中萎缩，以维持正常的呼吸作用。因此，研究呼吸过程中处于压缩和扩张作用下的肺表面活性剂单层膜的结构特征和力学性能是肺表面活性剂生物物理学的核心问题。实验方法无法直接观察到单层膜在压缩过程中结构转变的动态行为和分子构象的细微变化，而全原子分子动力学模拟和MARTINI粗粒化模拟又在模拟体系的空间和时间尺度上受限，模拟大尺度的复杂生物体系仍然面临诸多挑战。

      本文中我们针对肺表面活性剂中主要成分二棕榈酰磷脂酰胆碱（dipalmitoylphosphatidylcholine, DPPC）和棕榈酰油酰磷脂酰胆碱（palmitoyloleoylphosphatidylcholine, POPC）在水气界面形成的磷脂单层膜，使用约束液滴表面张力测量仪（constrained drop surfactometry, CDS）对其进行了实验测量并建立了介观尺度的单层膜粗粒化模型，并采用多体耗散粒子动力学（multi-body dissipative particle dynamics, MDPD）方法对单层膜进行了全面的模拟研究。对于饱和的DPPC和不饱和的POPC这两种磷脂分子形成的单层膜，我们发现环境温度的不同是其压缩等温线形态以及位置改变的主要因素，而实验所使用的面积压缩速率则对等温线没有显著的影响。对于DPPC单层膜，所处温度的升高导致其表面压力增加，等温线上的相共存区域也将会变得更窄。而对于具有简单相行为的POPC单层膜，温度的升高只降低了其表面张力。最后我们发现混合的DPPC/POPC单层膜表现出中间相的行为，并随着两种PC含量占比的改变而显示出不同的变化趋势。

        基于自上而下的粗粒化方案，我们建立了DPPC和POPC的单层膜介观模型，并使用MDPD模拟得到了与全原子模拟和CDS实验相似的表面压力-面积等温线，观察到与表面压力相关的单层膜形态，得到了处于不同相区域的单层膜。此外，面积压缩模量，脂质尾部的序参数、单层膜的厚度以及珠子的密度分布均与之前的模拟和实验结果在定量上相符，还捕获了混合DPPC/POPC单层膜的表面压力-面积等温线随着两种磷脂含量占比的敏感变化。综上，我们所建立的MDPD单层膜模型在计算效率更高的前提下，能够在更大时空尺度上准确模拟复杂单层膜的结构特征、力学特性和界面行为。

      Lung surfactant is a biologically active substance adsorbed on the surface of alveoli, mainly composed of lipids and proteins. It is assembled by alveolar type II cells and secreted on the surface of the alveoli to form a liquid film. The surfactant film covering the surface of the alveoli is the first barrier to prevent pathogens or particles from inhaling the human body. In addition, it could reduce the surface tension on the water-air interface of the alveoli and prevent the alveoli from shrinking at the end of the breathing to maintain normal respiration. Therefore, studying the structure features and mechanical properties of the lipid monolayer during compression and expansion is central to elucidating the biophysics of lung surfactant. The experimental method cannot directly observe the dynamic behavior of the structural transformation and the subtle changes in the molecular conformation of the monolayer during the compression process. However, the all-atom molecular dynamics simulation and the MARTINI coarse-grained simulation are limited in the space and time scale of the simulation system. Therefore, the simulation of large-scale complex biological systems still faces a lot of challenges.

   In this thesis, we focus on the phospholipid monolayer formed by dipalmitoylphosphatidylcholine (DPPC) and palmitoyloleoylphosphatidylcholine (POPC) at the air-water interface. We carried out the experimental measurements on the monolayer by using constrained drop surfactometry (CDS), and established a coarse-grained model of the mesoscopic monolayers. Then we conducted a comprehensive simulation study on the monolayer using the multi-body dissipative particle dynamics (MDPD). For the monolayer formed by the two types of phospholipid molecules, saturated DPPC and unsaturated POPC, we found that the difference in ambient temperature is the main factor in the change of the compression isotherm shape and position, while the compression rate of monolayer used in the experiment has no significant effect on the isotherm. For the DPPC monolayer, the increase in temperature will increase its surface pressure, and the phase coexistence area on the isotherm become narrower. For the POPC monolayer with simple phase behavior, the increase in temperature only reduces its surface tension. Finally, it is found that the mixed DPPC/POPC monolayer exhibited mesophase behavior, and showed different trends as the proportions of the two PCs changed.

       A coarse-grained model of the lipid monolayer was established for mesoscopic simulation based on the top-down coarse-grained scheme. Using the MDPD method, we obtained the surface pressure-area isotherms including different phase regions, which is similar to the all-atom simulation and CDS experiment. And the monolayer morphology related to the surface pressure was observed in our monolayer models. In addition, the area compression modulus, the order parameter of the lipid tail, the thickness of the lipid monolayer and the density distribution of the beads are quantitatively consistent with the previous simulation and experimental results. The sensitive changes in the surface pressure-area isotherm of the mixed DPPC/POPC monolayer with varying mixing ratios was captured which indicating that our mesoscopic model has a significant improvement over the previous coarse-grained model. In summary, the MDPD monolayer model we established is capable of accurately and quantitatively simulating the structural features, mechanical properties and interface behavior of complex lipid monolayers with larger spatial and time scales  under the premise of higher computational efficiency.