Functional analysis of genes selected in each step.

(A) To determine the statistical significance of selected genes, we counted the number of genes identified in each step of our analysis. (B) Utilizing a set of 93 genes that are implicated in GBMs as of Aceview we calculated the statistical significance of the overlap (numbers in parentheses) with a hypergeometric distribution. We found that the significance increased, applying the steps in our approach. (C) Calculating p-values with a modified Fisher's exact test, we obtained a similar result for a set of glioma genes as of DAVID as well.

(A) 为确定筛选基因的统计学显著性，我们统计了本研究分析各环节中鉴定得到的基因数量。(B) 依托截至Aceview数据库收录的、与胶质母细胞瘤（GBMs）相关的93个基因集，我们通过超几何分布（hypergeometric distribution）计算了该基因集与本研究分析结果的重叠显著性（括号内为重叠基因数）。研究发现，采用本研究的分析流程后，显著性水平呈上升趋势。(C) 采用修正后的费希尔精确检验（modified Fisher's exact test）计算p值，针对截至DAVID数据库收录的胶质瘤基因集，我们也得到了相似的实验结果。