数据库代码公开：TCGA进行癌与癌旁的配对差异分析

#step1 提取出TCGA列名前12位相同的样本
#step2 TCGA列名第14位是1代表是正常组或癌旁，新增列tissue命名为"Paired_Normal",0代表肿瘤，新增列tissue命名为"Paired_Tumor"
#step3 从data中构造临时文件，每一次提取一种肿瘤进行Wilcoxon Signed Rank Tests，参数是paired = T
wilcox_test_p <- sig <- c()
for(i in CancerType) {
  tmp <- data[which(data$CancerType == i),]
  tumor <- tmp[which(tmp$tissue == "Paired_Tumor"),]
  normal <- tmp[which(tmp$tissue == "Paired_Normal"),]
  wp <- c(wp, wilcox.test(tumor$exp,
                          normal$exp,
                          paired = T)$p.value)
  sig <- c(sig, as.character(cut(wilcox.test(t$exp,
                                             n$exp,
                                             paired = T)$p.value,
                                 c(0,0.0001,0.001,0.01,0.05,1),
                                 c("****","***","**","*","")
                                 )
                             )
           )
}
names(wilcox_test_p) <- names(sig) <- unique(data$CancerType)
wilcox_test_p