GeneBank accession 2 Entrez gene id

Question

[盛夏]

I have a list id of GenBank accession like this:

NM_176890.1
BX101169.1
BM716958.1
NM_173652.1
AI808125.1
AA972198.1
AA235299.1
NM_003444.1
NM_012464.3

and I want to change it to entrez gene id, I tried DAVID, but failed, and BioMart, and others failed to. Can anyone advise some ways to me to finish this aim?

Thanks very much.

[Sean Davis]

Note that most genbank accession numbers will not be directly linked to an Entrez Gene ID. That said, when you tried to do the conversions using online tools, what "failed"?

[盛夏]

you mean like the genebank accession AA235299 is a EST, so it is not a gene, can't be mapped?

[nivi1972]

If you open these ID's in GenBank,the database itself contain link to entrez. For eg. open the site http://www.ncbi.nlm.nih.gov/genbank/. Paste your query like NM_176890 in the search text box and you will get the entire information (http://www.ncbi.nlm.nih.gov/nuccore/NM_176890 ). Here if you search Geneid it will highlight it like this "GeneID:259296"

[盛夏]

but I need to change a lot of the GenBank accessions to entrez gen ids once a time. I cant query it one by one

Answer 1

[Guangchuang Yu]

you can use clusterProfiler to convert gene IDs. see http://www.bioconductor.org/packages/3.2/bioc/vignettes/clusterProfiler/inst/doc/clusterProfiler.html#bitr-biological-id-translator

> acc=read.table("acc.txt", header=F, stringsAsFactors=F)
> acc
           V1
1 NM_176890.1
2  BX101169.1
3  BM716958.1
4 NM_173652.1
5  AI808125.1
6  AA972198.1
7  AA235299.1
8 NM_003444.1
9 NM_012464.3
> acc=gsub("\\.\\d", "", acc[,1])
> acc
[1] "NM_176890" "BX101169"  "BM716958"  "NM_173652" "AI808125"  "AA972198"
[7] "AA235299"  "NM_003444" "NM_012464"
> require(clusterProfiler)
> bitr(acc, fromType="ACCNUM", toType="ENTREZID", annoDb="org.Hs.eg.db")
     ACCNUM ENTREZID
1 NM_176890   259296
4 NM_173652   285154
8 NM_003444     7710
9 NM_012464     7092
Warning message:
In bitr(acc, fromType = "ACCNUM", toType = "ENTREZID", annoDb = "org.Hs.eg.db") :
  55.56% of input gene IDs are fail to map...
>

Some of the accession numbers your listed are not genes, and can't be mapped to entrez gene ID.

You can search them to verify, for example the AA235299.1 can be found in http://www.ncbi.nlm.nih.gov/nucest/1859736/.

[盛夏]

ok, I know it thanks very much~

[jfo]

Hi! How do we use this for accessions representing multiple organisms, especially non-model ones? for example: XP_021371371.1 and PIK53657.1

[Dast0840]

At present, when using annoDb it raises an error: Error in bitr(acc, fromType = "ACCNUM", toType = "ENTREZID", annoDb = "org.Hs.eg.db") : unused argument (annoDb = org.Hs.eg.db), therefore, it should be replaced with OrgDb = org.Hs.eg.db.

Answer 2

[pevsner]

You can copy and paste these 9 accession numbers into NCBI Nucleotide and see that five are from the EST database. Of the four that are RefSeq accessions, two have been updated and two have had their records removed.

[盛夏]

then, what should I to deal with the accession that has already removed or from the EST? just abandon it?

[Sean Davis]

For ESTs, though it is a bit old-fashioned, you could look at UniGene, which maps ESTs to "clusters". Many of those clusters are, then, mapped to Entrez Gene IDs.