循環繪圖和多分組多基因（變量）差異展示

經常遇到‘一個循環繪製每一個差異基因在腫瘤和正常的表達差異’和‘需要在一張圖中展示多個差異基因在腫瘤和正常組的表達分佈’需求。如下列兩張圖所示：

• 首選應得到data如下列所示：

然後循環作圖：

list <- names(data)[2:11]
list
plot_list = list() 
for (i in 1:3) { 
  p = ggboxplot(data, x='group',  y= as.character(list[i]), 
                bxp.errorbar = T, color = 'group',
                outlier.shape = NA,  add = "point",
                palette =c("#00AFBB","#FC4E07") )+stat_compare_means()
  plot_list[[i]] = p } 
# Another option: create pdf where each page is a separate plot. 
pdf("plots.pdf") 
for (i in 1:10) {print(plot_list[[i]]) } 
dev.off() 

• 將寬數據data轉換成長數據data1

#=======================================================
#一張圖
#=======================================================

library(reshape2)
library(tidyr)
colnames(data)
names(data)[2]
names(data)[11]
#使用data數據框中非subtype的第一個列名和最後一個列名
data1 <- gather(data, gene, count, BCL11A:ZCCHC7 )
head(data1)
p <- ggboxplot(data1, x = "gene", y = "count",size=0.5, bxp.errorbar = T, 
               color = "group",  palette =c( "#FC4E07", "#00AFBB"),
               add.params = list(size=0.5), outlier.shape = NA, 
               add = "point") + 
stat_compare_means(aes(group = group), label = "p.format")
p
pdf(file = 'expr_subtype.pdf', height = 4,width = 14)
p
dev.off()