TCGA_不同gleason评分的 gene 表达

library("stringr")
library("dplyr")
library("ggplot2")
library("readr")

加载数据（将第1列转为行名）：

if(T){
  dataFilt=read_csv(file = 'dataFilt.csv') 
  dataFilt=as.data.frame(dataFilt) 
  rownames(dataFilt)=dataFilt[,1] 
  dataFilt=dataFilt[,-1]
  dataFilt=log(dataFilt+1) #取对数
} #测序数据

if(T){
  data_cl=read_csv(file = 'data_cl_index.csv') 
  data_cl=as.data.frame(data_cl) 
  rownames(data_cl)=data_cl[,1] 
  data_cl=data_cl[,-1]
} #临床数据

一、合并临床和测序数据

1 更改dataFilt使其能与data_cl匹配

dataFilt_t=t(dataFilt) #转置

dataFilt_t=as.data.frame(dataFilt_t)
dataFilt_t <- add_rownames(dataFilt_t, "Sample") #add_rownames()使row.names变为第一列（data.frame）
colnames(data_cl)[1]="Sample" 

dataFilt_t$Sample=str_sub(dataFilt_t$Sample,1,12) #取病人编号

2 合并（便于后面任意提取基因）

data_m <- inner_join(data_cl,dataFilt_t, by= "Sample") #inner_join按by指定合并数据

二、获取包含“Gene表达”、“对应gleason评分”的data.frame

1 从data_m中提取Gene和临床信息

Gene='CHAC1'
gene_cl<- data_m %>% dplyr:: select( 1:ncol(data_cl),all_of( Gene ))

2 gene_cl加上临床信息“gleason_sum”

t_gleason1=as.numeric(substr(data_m$primary_gleason_grade,9,9)) #substr()用于从“Pattern 3”中提取数字，“9”为数字对于字符序
t_gleason2=as.numeric(substr(data_m$secondary_gleason_grade,9,9))
t_gleason_sum=t_gleason1+t_gleason2

gene_cl=cbind(gene_cl,t_gleason_sum)

3 gene_cl中提取Gene和“gleason_sum”

gene_gleason=select(gene_cl,'Sample',Gene,'t_gleason_sum') #后三个元素为列名

 

 

三、ggplot作图

gene_gleason$t_gleason_sum<- as.factor(gene_gleason$t_gleason_sum) #离散变量转factor

p = ggplot(gene_gleason,aes(x = t_gleason_sum, y = gene_gleason[,Gene],color=t_gleason_sum)) +
  geom_boxplot() + 
  xlab("gleason score") +
  ylab(Gene) +
  theme_bw() #白色背景
p