针对复杂疾病治愈率低且容易复发的现状，如何深层次精确地挖掘复杂疾病的致病通路是医学界亟待解决的重要科学问题之一。本课题拟结合三维基因组Hi-C数据及全基因组范围内整个染色质的空间位置关系，采用SNPs富集法和CRISPR-Cas9池筛选法检测染色质中的调控元件；结合RNA-seq、ChIP-seq等数据及Hi-C数据的锚区域信息和目标区域信息，使用序列同源模型挖掘被调控的目标元件；利用互信息和条件互信息计算统计独立性，并运用路径匹配算法剔除假阳性调控关系，进而构建生物分子调控网络；解析调控元件的动态变化规律及特性，并构建调控元件-目标元件-癌症的三层异构网络模型；设计适用于三层异构网络模型聚类算法，挖掘调控层面的癌症致病通路。研究结果将有助于精准靶标调控元件，从调控层面调整基因的表达和转录，实现对致病基因的正常调控，并为复杂疾病的诊治、药物靶标和研制提供新视角和理论依据。

In view of the low cure rate and easy recurrence of complex diseases, it is one of the important scientific issues how to deeply and accurately excavate the pathogenesis of complex diseases in the medical field. This study will use the Hi-C data of the three-dimensional genome and the spatial position of the whole chromatin in the whole genome, will employ SNPs enrichment analysis and CRISPR-Cas9 pool screening method to detect regulatory elements in the chromatin, mine the controlled target elements by using sequence homology models based on RNA-seq, ChIP-seq and Hi-C data anchor region information and target region information, construct a biomolecular regulatory network by using mutual information and conditional mutual information to calculate statistical independence and using path matching algorithm to eliminate false positive regulatory relationship, analyze the dynamic change rules and characteristics of regulatory elements , and construct a three-layer heterogeneous network model of regulatory elements-target elements-cancer, as well as design clustering algorithm for three-layer heterogeneous network models to explore the pathogenic pathways in cancer at the regulatory level. The results will help to precisely target regulatory elements, regulate gene expression and transcription from the regulatory level, so as to achieve normal regulation of pathogenic genes, and provide a new perspective and theoretical basis for the diagnosis and treatment, drug targets and development of complex diseases.