差异基因富集分析（R语言——GO&KEGG&GSEA）

差异基因富集分析（R语言——GO&KEGG&GSEA）

引用

CSDN

1.

https://m.blog.csdn.net/m0_74823892/article/details/145804059

本文主要介绍了使用R语言进行差异基因富集分析的具体步骤，包括GO、KEGG和GSEA三种富集分析方法。文章内容详细，包含了具体的R代码示例，适合有一定生物信息学背景的读者参考学习。

1. 准备差异基因集

在进行通路富集分析之前，我们需要准备差异基因集。通常我们会将差异基因分为上调基因和下调基因分别进行分析。下面是一些示例代码：

##载入所需R包
library(readxl)
library(DOSE)
library(org.Hs.eg.db)
library(topGO)
library(pathview)
library(ggplot2)
library(GSEABase)
library(limma) 
library(clusterProfiler)
library(enrichplot)

##edger
edger_diff <- diff_gene_Group
edger_diff_up <- rownames(edger_diff[which(edger_diff$logFC > 0.584962501),])
edger_diff_down <- rownames(edger_diff[which(edger_diff$logFC < -0.584962501),])

##deseq2
deseq2_diff <- diff_gene_Group2
deseq2_diff_up <- rownames(deseq2_diff[which(deseq2_diff$log2FoldChange > 0.584962501),])
deseq2_diff_down <- rownames(deseq2_diff[which(deseq2_diff$log2FoldChange < -0.584962501),])

##将差异基因集保存为一个list
gene_diff_edger_deseq2 <- list()
gene_diff_edger_deseq2[["edger_diff_up"]] <- edger_diff_up
gene_diff_edger_deseq2[["edger_diff_down"]] <- edger_diff_down
gene_diff_edger_deseq2[["deseq2_diff_up"]] <- deseq2_diff_up
gene_diff_edger_deseq2[["deseq2_diff_down"]] <- deseq2_diff_down

2. 进行通路富集分析

这里主要介绍普通的GO、KEGG和GSEA的简单富集。筛选显著富集通路的筛选条件也是根据自己的需求决定，一般是矫正后P值小于0.05。

for (i in 1:length(gene_diff_edger_deseq2)){
  keytypes(org.Hs.eg.db)

  entrezid_all = mapIds(x = org.Hs.eg.db,  
                        keys = gene_diff_edger_deseq2[[i]], 
                        keytype = "SYMBOL", #输入数据的类型
                        column = "ENTREZID")#输出数据的类型
  entrezid_all  = na.omit(entrezid_all)  #na省略entrezid_all中不是一一对应的数据情况
  entrezid_all = data.frame(entrezid_all) 

  ##GO富集##
  GO_enrich = enrichGO(gene = entrezid_all[,1],
                       OrgDb = org.Hs.eg.db, 
                       keyType = "ENTREZID", #输入数据的类型
                       ont = "ALL", #可以输入CC/MF/BP/ALL
                       #universe = 背景数据集我没用到它。
                       pvalueCutoff = 1,qvalueCutoff = 1, #表示筛选的阈值，阈值设置太严格可导致筛选不到基因。可指定 1 以输出全部
                       readable = T) #是否将基因ID映射到基因名称。

  GO_enrich_data  = data.frame(GO_enrich)
  write.csv(GO_enrich_data,paste('GO_enrich_',names(gene_diff_edger_deseq2)[i], '.csv', sep = ""))
  GO_enrich_data <- GO_enrich_data[which(GO_enrich_data$p.adjust < 0.05),]
  write.csv(GO_enrich_data,paste('GO_enrich_',names(gene_diff_edger_deseq2)[i], '_filter.csv', sep = ""))

  ###KEGG富集分析###
  KEGG_enrich = enrichKEGG(gene = entrezid_all[,1], #即待富集的基因列表
                           keyType = "kegg",
                           pAdjustMethod = 'fdr',  #指定p值校正方法
                           organism= "human",  #hsa，可根据你自己要研究的物种更改，可在https://www.kegg.jp/brite/br08611中寻找
                           qvalueCutoff = 1, #指定 p 值阈值（可指定 1 以输出全部）
                           pvalueCutoff=1) #指定 q 值阈值（可指定 1 以输出全部）
  KEGG_enrich_data  = data.frame(KEGG_enrich)
  write.csv(KEGG_enrich_data, paste('KEGG_enrich_',names(gene_diff_edger_deseq2)[i], '.csv', sep = ""))
  KEGG_enrich_data <- KEGG_enrich_data[which(KEGG_enrich_data$p.adjust < 0.05),]
  write.csv(KEGG_enrich_data, paste('KEGG_enrich_',names(gene_diff_edger_deseq2)[i], '_filter.csv', sep = ""))
}

3. 通路富集情况可视化

这里只介绍一种简单的气泡图，当然还有其他的自己去了解吧。

##GO&KEGG富集BPCCMFKEGG分面绘图需要分开处理一下，富集结果里的ONTOLOGYL列修改
GO_enrich_data_BP <- subset(GO_enrich_data, subset = GO_enrich_data$ONTOLOGY == "BP")
GO_enrich_data_CC <- subset(GO_enrich_data, subset = GO_enrich_data$ONTOLOGY == "CC")
GO_enrich_data_MF <- subset(GO_enrich_data, subset = GO_enrich_data$ONTOLOGY == "MF")
##提取GO富集BPCCMF的top5
GO_enrich_data_filter <- rbind(GO_enrich_data_BP[1:5,], GO_enrich_data_CC[1:5,], GO_enrich_data_MF[1:5,])
##重新整合进富集结果
GO_enrich@result <- GO_enrich_data_filter
##处理KEGG富集结果
KEGG_enrich@result <- KEGG_enrich_data
ncol(KEGG_enrich@result)
KEGG_enrich@result$ONTOLOGY <- "KEGG"
KEGG_enrich@result <- KEGG_enrich@result[,c(10,1:9)]
##整合GO KEGG富集结果
ego_GO_KEGG <- GO_enrich
ego_GO_KEGG@result <- rbind(ego_GO_KEGG@result, KEGG_enrich@result[1:5,])
ego_GO_KEGG@result$ONTOLOGY <- factor(ego_GO_KEGG@result$ONTOLOGY, levels = c("BP", "CC", "MF","KEGG"))##规定分组顺序
##简单画图
pdf("edger_diff_up_dotplot.pdf", width = 7, height = 7)
dotplot(ego_GO_KEGG, split = "ONTOLOGY", title="UP-GO&KEGG", label_format = 60, color = "pvalue") + 
  facet_grid(ONTOLOGY~., scale = "free_y")+
  theme(plot.title = element_text(hjust = 0.5, size = 13, face = "bold"), axis.text.x = element_text(angle = 90, hjust = 1))
dev.off()

4. 气泡图如图所示

做了些处理，真实图片，左侧pathway是跟后面气泡一一对应的，当然还有其他可视化方式那就需要各位自己去探索了，谢谢！

5. GSEA富集分析

这里也是做一下简单的GSEA

##GSEA官方网站下载背景gmt文件并读入
geneset <- list()
geneset[["c2_cp"]] <- read.gmt("c2.cp.v2023.2.Hs.symbols.gmt")
geneset[["c2_cp_kegg_legacy"]] <- read.gmt("c2.cp.kegg_legacy.v2023.2.Hs.symbols.gmt")
![](https://wy-static.wenxiaobai.com/chat-rag-image/4693815346210228602)
geneset[["c2_cp_kegg_medicus"]] <- read.gmt("c2.cp.kegg_medicus.v2023.2.Hs.symbols.gmt")
geneset[["c2_cp_reactome"]] <- read.gmt("c2.cp.reactome.v2023.2.Hs.symbols.gmt")
geneset[["c3_tft"]] <- read.gmt("c3.tft.v2023.2.Hs.symbols.gmt")
geneset[["c4_cm"]] <- read.gmt("c4.cm.v2023.2.Hs.symbols.gmt")
geneset[["c5_go_bp"]] <- read.gmt("c5.go.bp.v2023.2.Hs.symbols.gmt")
geneset[["c5_go_cc"]] <- read.gmt("c5.go.cc.v2023.2.Hs.symbols.gmt")
geneset[["c5_go_mf"]] <- read.gmt("c5.go.mf.v2023.2.Hs.symbols.gmt")
geneset[["c6"]] <- read.gmt("c6.all.v2023.2.Hs.symbols.gmt")
geneset[["c7"]] <- read.gmt("c7.all.v2023.2.Hs.symbols.gmt")

##进行GSEA富集分析，这里也是写了个循环
gsea_results <- list()
for (i in names(gene_diff)){
  geneList <- gene_diff[[i]]$logFC
  names(geneList) <- toupper(rownames(gene_diff[[i]]))
  geneList <- sort(geneList,decreasing = T)
  for (j in names(geneset)){
    listnames <- paste(i,j,sep = "_")
    gsea_results[[listnames]] <- GSEA(geneList = geneList,
                         TERM2GENE = geneset[[j]],
                         verbose = F,
                         pvalueCutoff = 0.1,
                         pAdjustMethod = "none",
                         eps=0)
  }
}

##批量绘图，注意这里如果有空富集通路，会报错！
for (j in 1:nrow(gsea_results[[i]]@result)) {
    p <- gseaplot2(x=gsea_results[[i]],geneSetID=gsea_results[[i]]@result$ID[j], title = 
    gsea_results[[i]]@result$ID[j]) 

    pdf(paste(paste(names(gsea_results)[i], gsea_results[[i]]@result$ID[j], sep = 
    "_"),".pdf",sep = ""))
    print(p)
    dev.off()
  }

6. GSEA富集最简单图形如下

分享到此结束了，希望对大家有所帮助。