（1）分析单个基因在TCGA多个类型肿瘤中的分布（正常/肿瘤）

• 通过TCGAbiolinks下载表达量数据集
• 注释将表达量矩阵转化成基于gene symbol的表达矩阵
• 绘图展示单个基因在tcga数据库泛癌中的分布
  第一步，下载表达矩阵

#=======================================================


#=======================================================


library(GenomicDataCommons)

setwd('D:\\SCIwork\\F16DMDmeta\\review\\TCGA')

rm(list=ls())


library(dplyr)

library(TCGAbiolinks)

library(dplyr)

library(DT)

library(SummarizedExperiment)

library(stringr)

#=======================================================


#=======================================================

cancer  <- TCGAbiolinks:::getGDCprojects()$project_id

cancer <- str_subset(cancer, "TCGA")

cancer <- sort(cancer)




for (i in 1:33) {
  cancer_select <- cancer[i]
  print(cancer_select)
  #下载rna-seq的counts数据
  suppressMessages({
    query <- GDCquery(
      project = cancer_select,
      data.category = "Transcriptome Profiling",
      data.type = "Gene Expression Quantification",
      workflow.type = "HTSeq - FPKM")  })
  
  
  if (is.null(query)){
    print(paste0("No FPKM data of solid normal tissue for ", cancer_select ))
  } else{
    
    GDCdownload(query, method = "api", 
                files.per.chunk = 150)
    expdat <- GDCprepare(query = query, save = TRUE,
                         save.filename = paste0(cancer_select,".rda"))
    count_matrix=assay(expdat)
    write.csv(count_matrix,
              file = paste( cancer_select,"Counts.csv",
                            sep = "-"))}}