Hi all, I am trying to do GO terms annotation using a set of interesting human genes(with Ensembl gene id) and background, they are from RNA-seq data. However, whether I using PANTHER or DAVID, there's always more than 20% gene ids can't be mapped and I want to save these unmapped ids. I have try using Ensembl to convert ID to other id types like entrez or hgnc id, however the recognized gene number is even less... I wonder if there's anything I can do to mapped all of them? Thanks a lot! Some unmapped genes for example:
ENSG00000004660 ENSG00000005249 ENSG00000005448 ENSG00000005486 ENSG00000005812 ENSG00000214548 ENSG00000217325 ENSG00000218416 ENSG00000223839
•
written
by
hellocita •
> x= 'ENSG00000004660 ENSG00000005249 ENSG00000005448 ENSG00000005486 ENSG00000005812 ENSG00000214548 ENSG00000217325 ENSG00000218416 ENSG00000223839'
> id
[1] "ENSG00000004660" "ENSG00000005249" "ENSG00000005448" "ENSG00000005486"
[5] "ENSG00000005812" "ENSG00000214548" "ENSG00000217325" "ENSG00000218416"
[9] "ENSG00000223839"
> require(clusterProfiler)
> bitr(id, OrgDb='org.Hs.eg.db', fromType="ENSEMBL", toType='ENTREZID')
'select()' returned 1:1 mapping between keys and columns
ENSEMBL ENTREZID
1 ENSG00000004660 84254
2 ENSG00000005249 5577
3 ENSG00000005448 84058
4 ENSG00000005486 57414
5 ENSG00000005812 26224
6 ENSG00000214548 55384
8 ENSG00000218416 100130449
9 ENSG00000223839 100133036
Warning message:
In bitr(id, OrgDb = "org.Hs.eg.db", fromType = "ENSEMBL", toType = "ENTREZID") :
11.11% of input gene IDs are fail to map...
You can try my package, clusterProfiler for ID conversion and enrichment analysis.
Thank you Guangchuang! This package is very useful! The only problem is I need to updated my R version:)
Log in to answer this question.
