Entering edit mode
4.8 years ago
Diego Morais
•
0
Is there a file mapping all genbank/accession to interpro?
I have BLAST outputs containing accessions like OES33925.1, and I'm converting these ids to interpro ids using the mygene R/bioconductor package.
To do so I'm using awk to get all the ids from the BLAST output, and then using the R package. But I have to repeat this process for each file. If I get a file mapping all the known genbank/accessions to interpro ids I could generate a hash table to annotate all my files.
The megan software provides a file containing such mapping, but it is available in a format that I can't read outside the software. Furthermore, these file was last updated on june 2018, and I would like to build a new one if a updated mapping is available.
If the species you're working with are represented in Ensembl then you could use the Ensembl API to generate such a file.
Thank you for your comment.
Unfortunately, some species of my samples are not represented in the Ensembl. I tried the biomaRt R/bioconductor package, and then I found the mygene API, which may be used on a species agnostic way.
I will try the interpro to uniprotKB mapping provided by the interpro, and the UniProt ID mapping tool to convert the genbank to UniProt.
There might be an approach but it's not gonna be a straightforward one, I'm afraid.
Interpro provides something called a local lookup service, this is basically what you described (requested) a kind of file which list the results of running Interpro over all nr proteins. Have a look here: LookUpService
this is a huge download file (talking several tens of Gb) and actually is a complete service you can run locally.
What I try to get to is that you could download it and there should be some file in it containing the requested data, but I can't really immediately tell you how feasible it is to get that file out of the whole system.