复杂疾病的多层网络建模与优化控制

Revealing the pathogenesis of complex diseases, the early diagnosis and identifying drug targets of complex diseases are the most challenging problems in the fields of biology and medicine. Network-based approaches have recently emerged as powerful tools for the study of complex diseases. This project aims at developing quantitative tools and computational methods to reveal the underlying mechanisms of complex diseases, identify key genes, disease modules and dynamical network biomarks and uncover potential drug targets of complex disease from a systems level under a new framework of multilayer network modeling. Specifically, based on the high-throughput multi-omics data, we first explore the construction of multilayer network-based dynamical models of regulatory networks of complex diseases by using the techniques of optimization and parallel algorithms. Then, we will integrate graph theories, statistical analysis and calculation of network topology information to identify key disease-associated genes and modules for complex diseases. Furthermore, we will study controllability of complex disease network using the techniques of tensor representation and decomposition based on under the framework of multilayer network modeling, which providing new strategies for identifying dynamical network biomarks and potential drug targets of complex diseases. This project will not only provide a new systems biology approach for revealing novel pathogenic mechanisms of complex diseases through which genes or proteins lead to disease but also aid the development of innovative intervention strategies for the prevention and treatment of complex diseases.

复杂疾病的致病机制研究、早期预测诊断和药物靶标筛选是目前生物医学领域最具有挑战性的问题，基于网络的系统生物学方法已成为研究复杂疾病的强有力的工具之一。本项目旨在多层网络模型的框架下发展定量的工具和方法来研究复杂疾病的致病机制、识别复杂疾病发生的动态网络生物标记物，探讨对复杂疾病早期预防和控制新策略。具体地，以高通量的多组学数据为基础，以优化算法和并行算法为技术支撑，建立复杂疾病调控网络的多层网络模型；整合图论、统计分析与网络拓扑信息的计算与方法等，探索复杂疾病的关键致病基因和疾病相关的重要功能模块；在多层网络框架下基于张量分解方法研究复杂疾病网络的可控性，以便从控制论这一全新的视角识别复杂疾病的动态网络标记物和药物靶标。研究成果将为解析复杂疾病的致病机制、早期预测和预防控制提供创新性思路和理论指导.

复杂疾病对人类健康造成了严重的威胁，由于复杂疾病是在多个基因、蛋白质及其环境等众多因素共同作用下发生的, 这些相互作用构成一个复杂的网络。因此用基于网络的计算系统生物学方法来研究复杂疾病已经成为当前的研究热点和前沿。本项目旨在大规模网络、多层网络模型的框架下发展定量的工具和方法来研究复杂疾病的致病机制、识别复杂疾病发生的动态网络生物标记物，探讨对复杂疾病早期预防和控制新策略。通过本项目的实施，发展了构建大规模调控网络和动态调控网络的新算法；建立了整合多尺度的高维维异质数据的时序多层网络的张量框架模型，利用张量代数的理论设计了定量指标和新算法来定量评估时序多层网络中节点的重要性，并应用到识别导致流感和心脏病等疾病的关键致病基因以及与疾病相关的重要功能模块；提出了“区域控制”的新概念, 从理论上证明了在系统满足利普希茨连续和耗散条件下复杂疾病动力系统的可控性 并推广到复杂系统的轨线控制；基于图论的方法设计了有效的算法去识别控制节点，并应用到癌症中在正常状态和疾病状态中切换的关键基因，发现了一些致病的关键标记物和药物靶标。这些研究成果为解析复杂疾病的致病机制、早期预测和预防控制提供创新性思路和理论指导.