本项目以蛋白互作位点为切入点，通过对植物-病原菌种间蛋白互作开展序列、结构和网络分析，研究植物-病原菌相互作用过程。首先，我们将开发一种基于序列共进化的蛋白互作位点预测算法，并将之与基于结构建模的预测方法整合，实现对拟南芥种内蛋白互作位点和拟南芥-病原菌种间互作位点的高覆盖度预测。然后，以拟南芥-丁香假单胞杆菌为模式系统，开展基于蛋白序列和结构的统计分析，总结种间蛋白互作位点的理化基础与进化特性。最后，我们将通过整合互作组与转录组数据，构建拟南芥分别响应共生菌（白粉病菌）与腐生菌（灰霉菌）的动态网络；根据蛋白互作关系与互作位点的竞争，分别找到病原菌蛋白在宿主中的靶蛋白与靶互作，并基于动态网络扰动方法推断关键的靶蛋白与靶互作；结合功能注释与网络拓扑，对共生菌与腐生菌选择靶蛋白与靶互作的策略进行系统比较。本项目的实施将从系统层次加深对植物-病原菌相互作用机制的理解，为培育抗病作物提供新的思路。

In the proposed project, we will investigate the plant-pathogen interaction process via the computational analysis of protein-protein interactions (PPIs) between them. To achieve this goal, we will focus on the large-scale PPI site data and characterize the properties of such inter-species PPIs at the sequence, structure and network levels. First, we will develop a novel protein interaction site predictor based on sequence co-evolution features. This predictor will be further combined with other state-of-the-art PPI site predictors that rely on structure modeling, in order to achieve high-coverage predictions of the Arabidopsis thaliana (Arabidopsis for short) intra-species PPI sites and the inter-species PPI sites between Arabidopsis and its pathogens. Moreover, a comprehensive analysis of the sequence and structure propensities of the inter-species PPI sites between Arabidopsis and Pseudomonas syringae (a model host-pathogen system) will be performed, so that we could summarize the physicochemical and evolutionary basis of these inter-species PPIs. Finally, we will study the inter-species PPIs at the network level. Two dynamic responsive networks will be constructed by integrating the Arabidopsis interactome and the transcriptional profile in response to the biotrophic pathogen Golovinomyces orontii and the necrotrophic pathogen Botrytis cinerea, respectively. While the pathogen-targeted protein can be easily assigned according to the PPI annotations, the pathogen-targeted interaction can also be easily deduced when it competes the same PPI site with inter-species interactions. Therefore, we could find key pathogen-targeted proteins or key pathogen-targeted interactions by evaluating the effect of their perturbation on the dynamic responsive networks. We will systematically analyze the network topology and function of these key pathogen-targeted proteins/interactions, and compare the target selection strategies of the biotrophic pathogen and the necrotrophic pathogen. We hope that the proposed project can strengthen our systems understanding of plant-pathogen interaction and provide helpful hints for the rational design of novel pathogen-resistant crops.