主成分分析方法在基因调控网中的应用研究

The Application of Principal Components Analysis on Gene Regulatory Networks

利用基因表达值推断基因调控网是生物学中的一个重要领域,在疾病治疗领域和药物研究领域具有很好的效用.状态空间模型是一种新型优化的模型,在基因调控网中给出系统理论的分析和研究,较动态贝叶斯网络模型在平行的精度下具有极低的运行时间.本论文在前人工作基础上,研究主成分分析方法在状态空间模型中的应用,通过对调控基因关系矩阵取和方法找出主导基因,从而优化之前的线性空间模型,得到较优的结果.

Gene Regulatory Networks(GRN) building from gene microarray data plays an important role in Biology, which also plays an important role both in healthy system and pharmaceutical field. The State Space Model(SSM) is a new optimization approach, which spends less time at the similar precision with the Dynamic Bayesian Networks(DBN) model. In this paper,based on the previous work, we concentrated on the study of the application of Principal Components Analysis(PCA) on GRN. We utilized the information of the core genes in the networks by the processing of the matrix of gene regulatory relationship, and gained better result.