基于引力场算法的人类复杂疾病基因调控网络构建与分析

Human complex diseases have a serious influence on human health and life, and it is the essence of the treatment and prevention of complex diseases to recognize the complex disease mechanism on the genetic level. Gene regulatory network construction and analysis of complex diseases have become the focus of bioinformatics research under a large number of achievements yielded by genome-wide association studies. In this project, we intend to research the gene regulatory network construction and analysis for complex diseases. Confronting with the limitations of current the gene regulatory network studies of complex diseases, we plan to research new gene expression clustering algorithms to provide accurate data for the study of complex disease gene regulatory networks. We will propose a novel solution space decompose strategy to increase the efficiency of network construction algorithm. We will propose novel mathematical models and optimization strategies to study gene regulatory network of complex diseases. We try to complete the construction algorithm with protein-protein interaction network. Parallel theories and techniques will be applied to improve the speedup. Thus, we can understand the pathogenesis of complex diseases in-depth, research and develop reconstruction and analysis software of genetic interaction network of human complex diseases. Our studies want to achieve automatic identification and analysis of genetic interaction network of complex diseases for development of systems biology, and it has an important theoretical significance and application value to explore the pathogenesis of complex diseases and drug design.

人类复杂疾病严重影响人们健康和生活，从遗传层面认知复杂疾病机理是治疗和预防复杂疾病的本质所在。在人类全基因研究突飞猛进的大背景下，复杂疾病基因调控网络的构建与分析已成为生物信息学研究领域的热点。本项目拟实现面向人类复杂疾病基因调控网络构建与分析研究，针对当前复杂疾病基因调控网络研究的局限，研究全新基因表达聚类算法，为进一步解析提供详细数据；研究缩小解空间技术，提高网络构建效率；提出新的数学模型和新的优化策略，实现复杂疾病基因调控网络构建；实现融合蛋白质网络的全基因关联研究；采用并行理论技术，提高时间加速比；深入理解复杂疾病的发病机理，研发复杂疾病遗传互作网络重构与分析软件。本项目的研究以期实现复杂疾病遗传互作网络自动识别与分析，推动系统生物学的研究发展，对于探究复杂疾病的发病机理和药物设计等具有重要的理论意义和应用价值。

人类复杂疾病严重影响人们健康和生活，从遗传层面认知复杂疾病机理是治疗和预防复杂疾病的本质所在。本项目旨在研究人类复杂疾病基因调控网络的构建与分析，针对当前复杂疾病基因调控网络研究的局限，研究全新基因调控网络构建与分析算法。本项目提出一种基于混合并行遗传算法和阈值限定法的新型基因调控网络构建算法，并将该方法应用在人类复杂疾病的皮肤黑色素瘤和2型糖尿病基因调控网络的构建上，计算结果与真实网络作对比，验证已有结果的正确性；提出线性概率关系预测法算法，用以预测基因之间相互作用网络，解决了传统基因调控网络构建算法准确度不高的问题；提出了一种新的预测蛋白质-蛋白质网络的方法，以避免以往预测方法的必需蛋白数目过小的限制，该算法关键是整合蛋白质数据，并在此基础构建网络，整合数据的五标准核心措施（点度中心性、中间中心度、接近中心性，特征向量中心，和子核心）集成到一个集成框架中；开发了一种新型的融合算法：微分顶点和微分边缘融合算法，并用该方法构建了共表达基因网络，用其鉴别正常和肿瘤结合微分环节的状态之间的共表达关系，通过这种方法，鉴定了1823个基因。然后用新算法gthc识别功能模块。本项目的研究以期实现复杂疾病遗传互作网络自动识别与分析，推动系统生物学的研究发展，对于探究复杂疾病的发病机理和药物设计等具有重要的理论意义和应用价值。