Convertion from Uniprot ID to Gene Symbol does not work well.

Question

[entropy]

This seems to be a basic process but somehow I could not find the best answer yet.

I am trying to convert Uniprot IDs into Gene Symbols. I run this code but about 1/4 of my list returns NA, Is there a better way to get the full conversion right?

uniprots[1:5] . # "A0QVH7" "A0R666" "A0SYQ0" "A3RGC1" "A5A4K8"

length(uniprots)  # 64102

z <- select(org.Hs.eg.db, uniprots, "SYMBOL", "UNIPROT")

dim(z) # 64320     2

length( which( is.na( z$SYMBOL ) ) ) # 15789

[GenoMax]

Verify ones that are not converting using UniProt's ID mapping tool (from UniProt ID to Gene name). They may either not be human genes or deprecated in recent releases.

[entropy]

Thanks. It looks there exist some. I just tried and got this:"1,370 out of 1,433 identifiers from UniProtKB AC/ID were successfully mapped to 1,092 Gene name IDs.".

For example I got this:

A7MBE0  SLC22A1

O08537  Esr2

and so on...

Actually, I uploaded 15K IDs but received only that much result, not sure why. I selected FROM: "UniProtKB AC/ID" , TO: "GENE NAME" in the drop down menu.

[GenoMax]

O08537 Esr2

That is a mouse protein.

A7MBE0 SLC22A1

This is a cow protein.

You are using human database/library in your R code above. So it predictably is not able to find these ID's.

[entropy]

Thanks. I think that answers my question. Is there a way to scan all at once instead of looping per database?

[GenoMax]

I think you are asking if this can be done via R. Perhaps someone else would suggest an appropriate package.

You could also do this by programmatically accessing UniProt's site.