R软件--GEO数据分析教程：差异性分析、富集分析（GO\KEGG\DO）

---

title: “GSE93798”
output: word_document

knitr::opts_chunk$set(echo = TRUE)

R Markdown

关于GSE93798的差异性分析、GO分析、KEGG分析、DO分析

rm(list = ls()) 


#RStudio控制台(console)中出现少量中文乱码的解决方法
Sys.setlocale("LC_ALL","Chinese")


# if (!requireNamespace("BiocManager", quietly = TRUE))
#   install.packages("BiocManager")
# BiocManager::available()
# 
# options(BioC_mirror="http://mirrors.ustc.edu.cn/bioc/")

library(BiocManager)

#BiocManager::install('limma')
library(limma)

#install.packages("ggplot2")
library(ggplot2)

exp<-read.table("GSE93798_series_matrix1.txt",header = T,row.names = 1,comment.char = "!")


annot<- data.table::fread("GPL570.annot.gz",skip ="ID",data.table = F)

b <- data.table::fread("GPL22945_family.soft.gz",skip ="ID",data.table = F)


annot<-annot[,c(1,2,3)]
#annot=subset(annot,annot$Gene.symbol!="")

exp$Gene=b$Symbol[match(rownames(exp),b$ID)]
#处理重复基因
exp<-aggregate(x=exp[,-ncol(exp)],by=list(exp$Gene),FUN ='mean')
colnames(exp)[1]="Gene"

write.csv(exp,"GSE93798_eXp_matrix.csv")

gset<-exp[,-1]
rownames(gset)<-exp[,1]

#样本分组情况
description<-factor(c(rep("Con_l",22),rep("IgA",20)))
design<-model.matrix(~description+0,gset)
colnames(design)<-c("con_l","IgA")

fit<-lmFit(gset,design)

#构建对比组别
cont.matrix<-makeContrasts(IgA-con_l,levels = design)
fit2<-contrasts.fit(fit,cont.matrix)

##差异分析
fit2<-eBayes(fit2)
tT