数据挖掘—KEGG通路图绘制和通路图标注上下调基因

原创

sheldor没耳朵

发布于 2026-02-09 15:40:24

2190

文章被收录于专栏：数据挖掘数据挖掘

数据挖掘—KEGG通路图绘制和通路图标注上下调基因

在做差异基因分析后，使用差异基因做KEGG富集时候，常常需要绘制某些KEGG通路图以及在通路图上标注上下调基因，这里我包装了个函数来快速的进行绘制

plot_kegg_pathview_tf(deg_shared,select_dt,"pathview")

主要传入3个参数，

第一个是用来做富集分析的差异基因列表，需要同时包含“Gene”,"logFC"两列，如

#演示用，logFC统一设置为1或者-1，通路图会根据logFC值的大小设定颜色的深浅
> deg_shared[1:2,]
   Gene logFC
1 KCNE4     1
2  ZIC1     1

第二个参数是常规使用clusterProfiler做KEGG的结果（挑选想绘图的一些通路即可），

target_gene <- fread("table+figure/target.csv",
                     data.table = F,header = T)
target_gene <- target_gene$x
head(target_gene)
#[1] "KCNE4"    "ZIC1"     "GRAMD1C"  "ARHGEF10" "TGFBR3"   "SOX9" 
#KEGG
kk <- KEGG_tf(target_gene)
kk_dt <- as.data.frame(kk@result)
kk_dt <- kk_dt[kk_dt$category != "Human Diseases", ]
kk_dt <- na.omit(kk_dt)
dim(kk_dt[kk_dt$pvalue < 0.05,])#9
#如本次以绘制满足pvalue < 0.05的9条通路为例
select_dt <- kk_dt[1:9,]
select_dt[1:2,]
# category              subcategory       ID           Description
# hsa04014 Environmental Information Processing      Signal transduction hsa04014 Ras signaling pathway
# hsa04140                   Cellular Processes Transport and catabolism hsa04140    Autophagy - animal
# GeneRatio  BgRatio      pvalue  p.adjust    qvalue                    geneID Count
# hsa04014      4/34 238/9561 0.009662153 0.2291311 0.2005078 RALBP1/PDGFRB/PLA2G5/EGFR     4
# hsa04140      3/34 169/9561 0.021737477 0.2847245 0.2491565          PPP2CA/VPS39/UBC     3

第3个参数即为输出路径

函数plot_kegg_pathview_tf.R

plot_kegg_pathview_tf <- function(deg_shared, select_dt, out_dir){
  
  library(clusterProfiler)
  library(org.Hs.eg.db)
  library(pathview)
  
  # 创建总目录
  if(!dir.exists(out_dir)){
    dir.create(out_dir, recursive = TRUE)
  }
  
  # SYMBOL -> ENTREZID
  gene_df <- bitr(deg_shared$Gene,
                  fromType = "SYMBOL",
                  toType = "ENTREZID",
                  OrgDb = org.Hs.eg.db)
  
  # 合并 logFC
  gene_df <- merge(gene_df, deg_shared,
                   by.x = "SYMBOL", by.y = "Gene")
  
  gene_list <- gene_df$logFC
  names(gene_list) <- gene_df$ENTREZID
  
  # 需要绘制的 KEGG ID
  ids <- select_dt$ID
  
  for(i in ids){
    
    pathway_folder <- file.path(out_dir, i)
    
    if(!dir.exists(pathway_folder)){
      dir.create(pathway_folder)
    }
    
    # 设置输出目录为该通路文件夹
    old_wd <- getwd()
    setwd(pathway_folder)
    
    pathview(gene.data  = gene_list,
             pathway.id = i,
             species    = "hsa",
             kegg.native = TRUE,
             same.layer  = FALSE)
    
    setwd(old_wd)
  }
  
  message("All pathway plots saved in: ", out_dir)
}

结果就是每个通路生成一个单独的文件夹，其中包含3个文件，分别是原始的KEGG通路图，标注上下调基因的通路图等