TCGA 数据下载及处理R语言脚本

以下是TCGA数据下载及处理的R语言脚本：

首先，需要安装以下R包：TCGAbiolinks，tidyverse，ggplot2，survival，survminer。

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

# 安装其他必要的包
install.packages(c("tidyverse", "ggplot2", "survival", "survminer"))

接下来，下载TCGA数据。例如，我们下载肺癌（LUSC）的RNA-seq和临床数据。

library(TCGAbiolinks)

# Set working directory
setwd("your_working_directory")

# Download RNA-seq data
query <- GDCquery(project = "TCGA-LUSC",
                  data.category = "Transcriptome Profiling",
                  data.type = "Gene Expression Quantification",
                  workflow.type = "HTSeq - FPKM",
                  legacy = TRUE,
                  platform = "Illumina HiSeq",
                  file.type = "results",
                  experimental.strategy = "RNA-Seq")
GDCdownload(query)

# Download clinical data
query <- GDCquery(project = "TCGA-LUSC",
                  data.category = "Clinical",
                  file.type = "xml")
GDCdownload(query)

接下来，我们可以将下载的RNA-seq数据导入到R中，并进行预处理。例如，我们可以通过log2转换标准化数据并删除低表达基因。

# Load RNA-seq data
LUSC_rnaseq <- GDCprepare(query, save = TRUE, save.filename = "LUSC_rnaseq")

# Log2 transformation and normalization
LUSC_rnaseq$log2 <- log2(LUSC_rnaseq$counts+1)
LUSC_rnaseq_norm <- normalizeBetweenArrays(LUSC_rnaseq$log2, method = "quantile")

# Remove low expressed genes
LUSC_rnaseq_norm_filter <- LUSC_rnaseq_norm[rowSums(LUSC_rnaseq_norm > 1) >= 20,]

最后，我们可以使用survival和survminer包对临床数据进行生存分析和可视化。

# Load clinical data
LUSC_clinical <- GDCprepare_clinic(query, clinical.info = "patient")

# Merge RNA-seq and clinical data
LUSC_data <- merge(LUSC_rnaseq_norm_filter, LUSC_clinical, by = "bcr_patient_barcode")

# Survival analysis
fit <- survfit(Surv(time, vital_status) ~ 1, data = LUSC_data)
ggsurvplot(fit, data = LUSC_data, pval = TRUE, conf.int = TRUE)

# Cox proportional hazards model
model <- coxph(Surv(time, vital_status) ~ gene1 + gene2 + gene3, data = LUSC_data)
summary(model)