图谱方法实现DLBCL信息基因的提取与分类

Using Mapping Method to Realize the Extraction and Classification of DLBCL Information Genes

对肿瘤信息基因的提取及基因表达谱数据的处理,是基因表达谱研究中至关重要的一步。基于图谱理论提出一种新的DLBCL信息基因提取方法。首先,对每一个基因在不同条件下的表达情况构图,使其便于利用图论知识挖掘规律;进而进行奇异值分解(SVD)获取图的谱信息,刻画出该基因的表达规律,根据图谱与理想模板的余弦夹角及距离的运算选取信息基因子集;最后,实验了两组公开数据集,实验结果验证了方法的可行性。

The extraction of tumor information genes and the processing of gene expression profile data is an important step in the study of gene expression profile.Based on map theory,a new DLBCL information gene extraction method is proposed.Firstly,the composition of the expression of each gene under different conditions makes it easy to exploit the law of graph theory.Then singular value decomposition(SVD)is carried out to obtain the spectral information of the graph,and the expression rule of this gene is described.A subset of information genes is selected according to the calculation of cosine angle and distance between the graph and the ideal template.Finally,two sets of open data sets were tested and the feasibility of the method was verified.