#install.packages('e1071')

#if (!requireNamespace("BiocManager", quietly = TRUE))
#    install.packages("BiocManager")
#BiocManager::install("preprocessCore")

#if (!requireNamespace("BiocManager", quietly = TRUE))
#    install.packages("BiocManager")
#BiocManager::install("limma")


library("limma")   
source("https://bioconductor.org/biocLite.R")

if (!requireNamespace("AnnotationDbi", quietly = TRUE)) BiocManager::install("AnnotationDbi")
library("AnnotationDbi")
#要自己準(zhǔn)備TCGA的樣本的表達(dá)文件exp
#BiocManager::install("org.Hs.eg.db")
#1、TCGA基因ID的轉(zhuǎn)換
library(stringr)
exp$Ensembl_ID <- rownames(exp)
exp$Ensembl_ID=str_sub(exp$Ensembl_ID,1,15)
library(clusterProfiler)
library(org.Hs.eg.db)
# 查看org.Hs.eg.db 包提供的轉(zhuǎn)換類型
keytypes(org.Hs.eg.db)
# 需要轉(zhuǎn)換的Ensembl_ID
Ensembl_ID <- exp$Ensembl_ID
# 采用bitr()函數(shù)進(jìn)行轉(zhuǎn)換
gene_symbol <- bitr(Ensembl_ID, fromType="ENSEMBL", toType=c("SYMBOL", "ENTREZID"), OrgDb="org.Hs.eg.db")
# 查看轉(zhuǎn)換的結(jié)果
head(gene_symbol)

gene_symbol= gene_symbol[match(exp$Ensembl_ID,gene_symbol$ENSEMBL),]#匹配到表達(dá)矩陣中
exp$SYMBOL <- gene_symbol$SYMBOL
#整理去重，把重復(fù)的取平均值

exp <- exp[,-(ncol(exp)-1)]
exp2<-avereps(exp,ID=exp$SYMBOL)
exp2 <- as.data.frame(exp2)
exp2 <- exp2[!is.na(exp2$SYMBOL),]#去除沒有匹配的
rownames(exp2) <- exp2$SYMBOL
exp <- exp2[,-ncol(exp2)]

#2、轉(zhuǎn)換為矩陣，分析需要數(shù)據(jù)
dimnames=list(rownames(exp),colnames(exp))
data=matrix(as.numeric(as.matrix(exp)),nrow=nrow(exp),dimnames=dimnames)

#3、去除正常樣本
data <- data[,group_list== "tumor"]


#3、再轉(zhuǎn)換數(shù)據(jù)
v <-voom(data, plot = F, save.plot = F)#校正文件，因?yàn)槲覀兿螺d的是轉(zhuǎn)錄的數(shù)據(jù)，必須轉(zhuǎn)化為芯片數(shù)據(jù)
out=v$E
out=rbind(ID=colnames(out),out)
write.table(out,file="uniq.symbol.txt",sep="\t",quote=F,col.names=F)   

#4、CIBERSORT
source("TMBimmune19.CIBERSORT.R")
results=CIBERSORT("ref.txt", "uniq.symbol.txt", perm=100, QN=TRUE)
#OK完美運(yùn)行
```r