前言:
微博参与话题 #给你四年时间你也学不会生信#
主要参考：GEO数据挖掘小尝试：（三）利用clusterProfiler进行富集分析
Y叔开发的R包clusterProfiler 的确是最好用的，没有之一,可参看为Y叔疯狂打call
1、安装clusterProfiler
> source("http://bioconductor.org/biocLite.R")
> biocLite('clusterProfiler')
2、ID转换
对于没有转换的gene ID，clusterProfiler也提供了bitr方法进行转换ID：
# example:
> x <- c("GPX3", "GLRX", "LBP", "CRYAB", "DEFB1", "HCLS1", "SOD2", "HSPA2",
"ORM1", "IGFBP1", "PTHLH", "GPC3", "IGFBP3","TOB1", "MITF", "NDRG1",
"NR1H4", "FGFR3", "PVR", "IL6", "PTPRM", "ERBB2", "NID2", "LAMB1",
"COMP", "PLS3", "MCAM", "SPP1", "LAMC1", "COL4A2", "COL4A1", "MYOC",
"ANXA4", "TFPI2", "CST6", "SLPI", "TIMP2", "CPM", "GGT1", "NNMT",
"MAL", "EEF1A2", "HGD", "TCN2", "CDA", "PCCA", "CRYM", "PDXK",
"STC1", "WARS", "HMOX1", "FXYD2", "RBP4", "SLC6A12", "KDELR3", "ITM2B")
> head(eg)
SYMBOL ENTREZID ENSEMBL
1 GPX3 2878 ENSG00000211445
2 GLRX 2745 ENSG00000173221
3 LBP 3929 ENSG00000129988
4 CRYAB 1410 ENSG00000109846
5 DEFB1 1672 ENSG00000164825
6 DEFB1 1672 ENSG00000284881
3、GO、KEGG富集分析
OrgDb库
enrichGO默认gene type是entrezID，但其他OrgDb支持的类型（ENSEMBLE，SYMBOL等）都可以通过参数keyType指定。
gene的ID type不一样，富集的结果也会有稍微的差异。
原gene list是entrezID，直接通过bitr转换成ensembl和symbol，分别做enrichGO。
发现entrezedID可能对应多个ENSEMBL的。
entrezedID和SYMBOL是一一对应的
3.1 GO富集分析
> library(clusterProfiler)
> genelist <- egcounts$ENTREZID
> head(genelist)
[1] 2878 2745 3929 1410 1672 1672
> # 检查是否有重复
> genelist[duplicated(genelist)]
[1] 1672 6648 1428
由于clusterProfiler富集分析推荐的输入文件是Entrez ID，因此这里提取的是Entrez ID，接下来就可以进行富集分析了：
> go <- enrichGO(genelist, OrgDb = org.Hs.eg.db, ont='ALL',pAdjustMethod = 'BH',pvalueCutoff = 0.05,
+ qvalueCutoff = 0.2,keyType = 'ENTREZID')
> head(go)
ONTOLOGY ID Description GeneRatio BgRatio pvalue
GO:0001503 BP GO:0001503 ossification 9/56 371/17653 2.299994e-06
GO:0030198 BP GO:0030198 extracellular matrix organization 8/56 341/17653 1.131006e-05
GO:0045926 BP GO:0045926 negative regulation of growth 7/56 244/17653 1.151994e-05
GO:0043062 BP GO:0043062 extracellular structure organization 8/56 395/17653 3.250950e-05
GO:0003416 BP GO:0003416 endochondral bone growth 3/56 22/17653 4.461661e-05
GO:0001649 BP GO:0001649 osteoblast differentiation 6/56 205/17653 4.563530e-05
p.adjust qvalue geneID Count
GO:0001503 0.003969790 0.003019045 5744/2719/3486/10140/2261/3569/6696/4653/6781 9
GO:0030198 0.006627806 0.005040480 22795/3912/1311/6696/3915/1284/1282/7077 8
GO:0045926 0.006627806 0.005040480 3929/1410/2719/2261/6696/978/5950 7
GO:0043062 0.013127755 0.009983723 22795/3912/1311/6696/3915/1284/1282/7077 8
GO:0003416 0.013127755 0.009983723 2261/1311/6781 3
GO:0001649 0.013127755 0.009983723 5744/3486/10140/3569/6696/4653 6
> dim(go)
[1] 68 10
> dim(go[go$ONTOLOGY=='BP',])
[1] 46 10
> dim(go[go$ONTOLOGY=='CC',])
[1] 15 10
> dim(go[go$ONTOLOGY=='MF',])
[1] 7 10
> # 进行简单的可视化
> barplot(go,showCategory=20,drop=T)
> dotplot(go,showCategory=50)
Rplot20.png
Rplot21.png
3.2 KEGG通路富集
KEGG通路富集函数用法与GO富集分析方法类似：
> kegg <- enrichKEGG(genelist, organism = 'hsa', keyType = 'kegg', pvalueCutoff = 0.05,pAdjustMethod = 'BH', minGSSize = 10,maxGSSize = 500,qvalueCutoff = 0.2,use_internal_data = FALSE)
> head(kegg)
ID Description GeneRatio BgRatio pvalue p.adjust
hsa04512 hsa04512 ECM-receptor interaction 6/35 82/7470 1.830700e-06 0.0002324989
hsa04151 hsa04151 PI3K-Akt signaling pathway 9/35 354/7470 2.568570e-05 0.0012944211
hsa04510 hsa04510 Focal adhesion 7/35 199/7470 3.057688e-05 0.0012944211
hsa05146 hsa05146 Amoebiasis 5/35 96/7470 7.552538e-05 0.0023979308
hsa05222 hsa05222 Small cell lung cancer 4/35 93/7470 8.773963e-04 0.0222858657
hsa05134 hsa05134 Legionellosis 3/35 55/7470 2.092473e-03 0.0442906717
qvalue geneID Count
hsa04512 0.0001984864 3912/1311/6696/3915/1284/1282 6
hsa04151 0.0011050591 2261/3569/2064/3912/1311/6696/3915/1284/1282 9
hsa04510 0.0011050591 2064/3912/1311/6696/3915/1284/1282 7
hsa05146 0.0020471353 3569/3912/3915/1284/1282 5
hsa05222 0.0190256458 3912/3915/1284/1282 4
hsa05134 0.0378113484 3306/3569/1917 3
> dim(kegg)
[1] 6 9
> # 简单可视化
> dotplot(kegg, showCategory=30)
Rplot22.png
后续的3.2 KEGG通路富集，由于我没有差异表达的文件，后续无法继续分析，待到下次有数据后再做下一步分析吧。