首先推荐Kobashttp://kobas.cbi.pku.edu.cn/kobas3/genelist/
上传你的蛋白序列 genelist，背景基因选相似物种就可以做kegg，kegg的结果画图参考
https://mp.weixin.qq.com/s/fwZ4rZOg8PdVb2ZE9wskoQ
data01<-read.table("你物种的ko结果.txt",header = T,sep="\t")
pathway=data.frame(data01[,c(1,4,6,7)])
pathway=head(pathway,n=20)
names(pathway)<-c("PATHWAY","Gene","Pvalue","Qvalue")
library(ggplot2)
ggplot(pathway,aes(Pvalue,PATHWAY))+
geom_point(aes(size=Gene,color=-1*log10(Qvalue)))+
scale_color_gradient(low="green",high = "red")+
labs(color=expression(-log[10](Qvalue)),size="Gene",
x="Pvalue",
y="Pathway name",
title="Top20 of Pathway enrichment")+
theme_bw()
write.table(pathway,"simpathway.txt",row.names=F,col.names=T,sep="\t")
image.png
image.png
我同时也去尝试kobas的go富集 但一直显示这个画面 不过kegg倒是挺快的
image.png
接下来可以用eggnog-mapper先注释然后构建自己的org.db 再用clusterProfiler包画，可以参考别人的做法 我做完感觉图还是不好看 ，构建库也是很慢
接下来做Go我使用eggnog-mapper先对我要的物种注释获得Go号
sim.emapper.annotations这是我的注释结果命名为sim
vi chang_go.py
#!/usr/bin/python
# -*- coding: utf-8 -*-
import sys, os
x = sys.argv[1]
file = open(x, "r")
lines = file.readlines()
for line in lines:
line=line.strip()
if line.startswith("#"):
continue
else:
tmp=line.split("\t")
if tmp[6] == "":
continue
else :
mystr=tmp[0]+"\t"+tmp[6]
print (mystr)
python chang_go.py sim >simout
sed 's/,/\t/g' simout >simout.txt
然后借用下某平台简单做一下
https://www.omicshare.com/tools/home/report/goenrich.html
把你的genelist和注释的结果一放就好了 结果会有类似这种图 和用R语言clusterProfiler包画的差不多吧 感觉应该可以这样做吧 请大家试试
image.png
6.7号更新
用eggnog-mapper画go富集分析代码 不用构建物种库 快速运行 节约时间
rm(list=ls())
setwd("D:")
BiocManager::install("stringr")
BiocManager::install("dplyr")
BiocManager::install("clusterProfiler")
library(stringr)
library(dplyr)
egg<-read.csv(".emapper.annotations",sep="\t",header=T)
egg[egg==""]<-NA
gene_ids <- egg$query_name
eggnog_lines_with_go <- egg$GOs!= ""
eggnog_lines_with_go
eggnog_annoations_go <- str_split(egg[eggnog_lines_with_go,]$GOs, ",")
gene_to_go <- data.frame(gene = rep(gene_ids[eggnog_lines_with_go],
times = sapply(eggnog_annoations_go, length)),
term = unlist(eggnog_annoations_go))
head(gene_to_go)
library(clusterProfiler)
gene_list <- read.table("ID")
gene_list <-as.vector(gene_list$V1)
term2gene<-gene_to_go[,c(2,1)]
term2gene=na.omit(term2gene)
df<-enricher(gene=gene_list,
pvalueCutoff = 0.05,
pAdjustMethod = "BH",
TERM2GENE = term2gene)
head(df)
barplot(df)
dotplot(df)
df<-as.data.frame(df)
dim(df)
df1<-go2term(df$ID)
dim(df1)
head(df1)
df$term<-df1$Term
df2<-go2ont(df$ID)
dim(df2)
head(df2)
df$Ont<-df2$Ontology
head(df)
df3<-df%>%
select(c("term","Ont","pvalue"))
#df3=subset(df,select=c("term","Ont","pvalue"))
head(df3)
library(ggplot2)
ggplot(df3,aes(x=term,y=-log10(pvalue)))+
geom_col(aes(fill=Ont))+
coord_flip()+labs(x="")+
theme_bw()
write.table(df3,"go_out",row.names=T,col.names=T,sep="\t")
Go号由一对多变为多对多 然后换agriGo做
library(tidyverse)
setwd("D:/kobas注释/go")
data01<-read.table("simout",header = F,sep="\t")
names(data01)<-c("gene","GO_ID")
data02<-data01 %>%
tidyr::separate_rows(GO_ID, sep = ",")
write.table(data02,"simgo.out",quote = FALSE,row.names=F,col.names=F,sep="\t")
library(ggplot2)
CPCOLS <- c("#8DA1CB", "#FD8D62", "#66C3A5")
p <- ggplot(data=data, aes(x=data$Term, y=LG10, fill=KEGG_A_class)) +
geom_bar(stat="identity", width=0.8) + coord_flip() +
scale_fill_manual(values = CPCOLS) + theme_bw() +
scale_x_discrete(labels=data) +
xlab("KGGG term") +
theme(axis.text=element_text(face = "bold", color="gray50")) +
labs(title = "The Most Enriched KEGG Terms")
###go富集 注释的分类图
rm(list=ls())
setwd("D:/")
library(openxlsx)
data<- read.xlsx("Functions enriched by Blast2GO.xlsx", sheet = 1)
###处理为ID Description GeneNumber type
data1=subset(data,select=c("GO.ID","GO.Name","Nr.test","GO.Category"))
library(ggplot2)
data1=data1[,c(2,3,4)]
colnames(data1)=c("name","number","type")
a=data1[data1$number>=10,]
a=a[order(a$type),]
ggplot(a,aes(x=factor(name,levels=unique(name)),y=number,fill=type))+geom_bar(stat="identity")+coord_flip()+
theme_bw() + ylab("Number") + xlab("Name")
###nr库比对注释图
首先diamond比对上nr库然后利用 UniProt2GO_annotate.py 脚本得到GO号
nohup diamond blastp -d nr.dmnd -q my.pep -o my.out -f 6 -p 10 --more-sensitive -e 1e-5 &
nohup python UniProt2GO_annotate.py ./idmapping/idmapping.tb.gz my.out my.go.out &
然后利用R语言
library(ggplot2)
options(stringsAsFactors = F)
library("dplyr")
library("stringr")
library(tidyverse)
library(GO.db)
male=read.table("my.go.out",header = F)
names(male)<-c("Gene","GOID")
male02<-male %>%
tidyr::separate_rows(GOID, sep = ",")
godb <- select(GO.db, keys(GO.db), columns(GO.db))
go_male02 <- male02 %>% left_join(godb)
go_annot=go_male02
go_annot=go_annot[,-1]
go_annot=go_annot[,-2]
as=as.data.frame(table(go_annot$GOID))
colnames(as)=c("GOID","GeneNumber")
go_annot2 <- go_annot %>% left_join(as)
go_annot=go_annot2 %>% distinct()#根据所有列删除重复的行（完全一样的观测值）
go_annot=go_annot[,c(1,3,2,4)]
colnames(go_annot)=c("ID","Description","GeneNumber","type")
go_enrich_df=go_annot
dat=go_enrich_df[,c(2,3,4)]
colnames(dat)=c("name","type","number")
a=dat[dat$number>=50,]
a=a[order(a$type),]
library(ggplot2)
ggplot(a,aes(x=factor(name,levels=unique(name)),y=number,fill=type))+geom_bar(stat="identity")+coord_flip()+
theme_bw() + ylab("Number") + xlab("Name")
###
KEGG注释图的绘制
首先上传蛋白质到KOBAS网站获取注释文件
import sys
input = sys.argv[1]
input_file = open(input, "r")
input_reads = input_file.read()
title = "query_name" + "\t" + "gene_name" + "\t" + "KO_number" + "\t" + "pathway"
print(title)
for lines in input_reads.split("Query"):
if "Pathway" in lines:
for line in lines.split("\n"):
if line.startswith(":"):
query_name = line.split("\t")[1]
if line.startswith("Gene:"):
gene_name = line.split("\t")[1]
if "KEGG" in line:
KO_number = str(line.split("\t")[3]).replace("pop", "")
pathway = line.split("\t")[1]
output_line = query_name + "\t" + gene_name + "\t" + KO_number + "\t" + pathway
print(output_line)
input_file.close()
···
得到
query_name gene_name KO_number pathway
Simonii00034466-RA pop:7475026 01100 Metabolic pathways
Simonii00034466-RA pop:7475026 01110 Biosynthesis of secondary metabolites
Simonii00034466-RA pop:7475026 00564 Glycerophospholipid metabolism
Simonii00034466-RA pop:7475026 00565 Ether lipid metabolism
Simonii00034466-RA pop:7475026 00562 Inositol phosphate metabolism
Simonii00036281-RA pop:7480371 03018 RNA degradation
Simonii00038638-RA pop:7472530 04075 Plant hormone signal transduction
Simonii00038638-RA pop:7472530 04016 MAPK signaling pathway - plant
Simonii00010134-RA pop:7478365 01100 Metabolic pathways
如上ko号 和注释通路
## KAAS在线注释
首先介绍怎样使用网站的在线注释功能。
进入KAAS：[https://www.genome.jp/kaas-bin/kaas_main](https://www.genome.jp/kaas-bin/kaas_main)