在做完KEGG通路富集之后，纷繁复杂的通路映入眼帘，略微有些杂乱无章，缺乏一个有序的排列，因此，本推送依据KEGG一级类目，将不同的通路归纳总结，分类排列，使结果更加规整，帮助后续进一步分析与研究

1. 准备工作

1.1 R包安装

library("clusterProfiler")
library("org.Hs.eg.db")
library("dplyr")
library("ggsci")
library("ggplot2")

ggsci的教程可以查看昨天的推送，现学现用~[ggsci | 一行代码拥有Sci级配色]

1.2 环境配置

关于为什么要进行环境配置可以查看既往推送[KEGG数据库报错及解决]，避免KEGG富集分析中报错。

options(clusterProfiler.download.method = "wininet")

1.3 文件准备

一共需要准备2个文件，一个是差异分析结果diffSig.xls，另一个是通路关系文件KEGG_relation.xls，本推送已整理好KEGG_relation.xls，文末有免费获取方式

# 1. 差异分析文件
diff <- read.table("./input/diffSig.xls",sep = "\t",header = T)
> head(diff)
     genes logFC
1  C4orf19  1.85
2  C3orf80  2.12
3  C2orf81  5.18
4 C17orf50  1.53
5 C11orf24  1.95
6  C9orf64  1.66

# 2. 通路关系文件
KEGG_relation <- read.table("./file/KEGG_relation.xls",sep = "\t",header = T)
> head(KEGG_relation)
              Class1                        terms
1 Cellular Processes            Adherens junction
2 Cellular Processes                    Apoptosis
3 Cellular Processes              Apoptosis - fly
4 Cellular Processes Apoptosis - multiple species
5 Cellular Processes           Autophagy - animal
6 Cellular Processes            Autophagy - other

2. 分析步骤

将差异分析结果（如1.3中的diffSig.xls）放入file文件夹中，全选运行code.R即可

library("clusterProfiler")
library("org.Hs.eg.db")
library("dplyr")
library("ggsci")
library("ggplot2")
options(clusterProfiler.download.method = "wininet")
diff <- read.table("./input/diffSig.xls",sep = "\t",header = T)
KEGG_relation <- read.table("./file/KEGG_relation.xls",sep = "\t",header = T)
genes=as.vector(diff[,1])
entrezIDs <- mget(genes, org.Hs.egSYMBOL2EG, ifnotfound=NA) %>%as.character(.)
kk <- enrichKEGG(gene = entrezIDs, organism = "hsa", pvalueCutoff = 0.05, qvalueCutoff = 2)
enrich_results <- subset(kk@result,pvalue < 0.05)
enrich_intersect <- inner_join(KEGG_relation,enrich_results,by=c("terms"="Description"))
enrich_intersect$Class1 <- factor(enrich_intersect$Class1,levels = unique(enrich_intersect$Class1))

KEGG_results <- ggplot(enrich_intersect, aes(x=terms, y=Count,fill=Class1)) +
  geom_bar(stat="identity", width=0.8) + # width可设置条形图宽度
  coord_flip() +
  xlab("") +
  theme_bw()+
  scale_fill_lancet(name = "Type")+
  scale_x_discrete(limits = factor(enrich_intersect$terms))+
  guides(fill = guide_legend(reverse=T))+
  theme_bw()+
  theme(panel.grid=element_blank())+
  theme(axis.title = element_text(size = 15,face = "bold",colour = "black"),
        axis.text = element_text(size=10,face = "bold",colour = "black"), 
        panel.border = element_rect(fill=NA,color="black", linewidth =3, linetype="solid"),
        axis.ticks = element_line(size = 2),
        legend.text = element_text(size = 10,face="bold"),
        legend.title = element_text(size = 10,face = "bold"))
KEGG_results
ggsave("KEGG_classification.PDF",plot = KEGG_results,width = 10,height = 8)

当然，此处输出的是所有P<0.05的结果，如果想要指定KEGG条目的话，可以查看推送[如何筛选感兴趣的GO和KEGG进行绘图~]

最终呈现效果

image

感谢观看，如果有用还请点赞，关注，在看，转发！