一种挖掘关键基因的新方法及其应用(英文)

A novel method of mining key genes and its application

近来,人们发现从疾病相关基因中寻找关键基因对疾病的诊断和治疗很重要。癌相关基因的网络是根据正常和患病的胶质瘤组织的基因表达谱建立。根据建立的基因网络和CIPHER方法,不同阈值下的正常和患病的胶质瘤表型网络被建立。根据已知的疾病和表型间的关联,另一组正常和患病的胶质瘤表型网络被建立。将两种方法建立的相应的表型网络进行比较,匹配度最大时对应的阈值及基因和表型网络被确定。在此基础上,通过打分方法得到了7个关键基因:DMBT1,ERBB2,NF2,PDGFB,AR,ARAF和TP53。文献查询发现其中5个基因与胶质瘤的形成和发展密切相关。剩下两个基因中的ARAF也间接地参与胶质瘤形成。因此,这两个基因可能在胶质瘤的形成中起重要作用。这一预测仍需要实验验证。

It is believed that finding key genes from disease-related genes is important for diagnosis and treatment of disease. In this work, networks of cancer-related genes are constructed based on their expression data in glioma tissues with and without cancer. Based on this, networks of glioma-related phenotypes for corresponding tissues are constructed using the method of CIPHER. Another group of networks of corresponding phenotypes for glioma tissues with and without canc- er are constructed based on the known correlations between diseases and phenotypes. When the matching degree between a phenotype network built by CIPHER under a certain threshold and that constructed by the known correlations reaches maximum for glioma tissues with and without cancer, the corresponding gene and phenotype networks under the threshold are obtained. Based on these determined gene and phenotype networks, seven key genes DMBT1, ERBB2, NF2, PDGFB, AR, ARAF and TP53 are given by scoring method. Our literature review shows that five of these seven genes are closely related to the formation and development of glioma, leaving AR and ARAF open. Furtheruaore, ARAF participates indirectly in the formation of glioma. Therefore, we predict the two open genes may play important roles in the formation of glioma. These predictions call for empirical studies.