基于多组学数据整合的不同时期胃癌动态调控机制研究

Significant difference of the 5-year survival rate of gastric cancer in different stages indicates the different gene regulatory mechanisms among different stages of gastric cancer. Deciphering the underlying mechanism can provide useful information for the accurate diagnosis and treatment of the cancer. Gene regulatory network is one of the powerful methods that understand the intracellular gene regulatory mechanism. Recently, the construction of gene regulatory network is mainly based on single-omic data. However, many factors are involved in the gene regulation mechanism and it is hard to ensure the reliability and specificity of the constructed gene regulatory network with single data resource. To overcome the problem, in this study we will integrate various omic data, such as gene expression, DNA methylation, H3K4me1/H3K27ac histone modification and chromatin accessibility data, to construct a more reliable and specific gastric cancer associated gene regulatory network. Through analyzing the mutual information among different types of biological features, we can mine several specific patterns of gene expression, epigenetic modification and infer related driver genes. These results can help us to get new insight into the underlying mechanism of tumorigenesis and potentially improve the tumor classification. This study will promote the establishment of the pipeline of network topologic analysis based on multi-omics data, which can be used for analyzing other complex diseases.

不同时期的胃癌5年存活率之间存在着显著的差异，对不同时期胃癌展开针对性的基因调控机制分析对胃癌的精准诊疗具有重要的意义。基因调控网络是研究细胞内基因调控机制的有效手段之一。目前针对基因调控网络的研究大多数是基于单一类型的组学数据，然而细胞内基因的调控机制受到多重因素的影响，因此单一数据源难以保证其特异性及可靠性。本项目拟通过整合多种组学数据，包括基因表达水平、DNA甲基化、H3K4me1/H3K27ac组蛋白修饰以及染色质可接近性数据，构建更可靠、更具特异性的胃癌基因调控网络，分别从时间及空间的尺度来研究其不同生物学特征的交互信息，挖掘与胃癌的形成和发展密切相关的特异性基因表达模式、表观遗传修饰信息以及相关的驱动基因，将有助于阐明胃癌的发生机制，提高胃癌分型的准确率。此项目的实施将推动多组学数据整合的网络拓扑分析框架的建立，并将其拓展至其它疾病的机制研究。

不同时期的胃癌5年存活率之间存在着显著的差异，对不同时期胃癌展开针对性的基因调控机制分析对胃癌的精准诊疗具有重要的意义。基因调控网络是研究细胞内基因调控机制的有效手段之一。目前针对基因调控网络的研究大多数是基于单一类型的组学数据，然而细胞内基因的调控机制受到多重因素的影响，因此单一数据源难以保证其特异性及可靠性。本项目通过整合不同类型的组学数据，构建了更可靠、更具有特异性的胃癌基因调控网络，并从时间尺度上揭示了肿瘤发展不同时期，基因调控网络的拓扑结构变化，并筛选了相关的肿瘤标志物。在研究的过程中，针对遇到的问题，如数据不平衡、DNA甲基化的图谱构建以及miRNA的作用，本项目也做出了适当的拓展性的研究。本项目的开展，是符合当前肿瘤研究领域中的多组学整合的理念，具有一定的研究意义，且相关的研究方法能够拓展到其它肿瘤领域的应用。