Interproscan sequence classification

Question

[drikaul]

Hi all,

I have InterProScan results for a bunch of protein queries, and I see comprehensive functional annotation and description for them. I was wondering if it's possible to get putative taxonomic classification for them as well, is interproscan designed to do that? Is there a way to retrieve any taxonomic annotation from the IPR/GO/Pfam annotation IDs?

Any advice is appreciated!

-Drishti

[lieven.sterck]

What specifically do you mean with 'putative taxonomic classification' ? Usually you run this kind of analysis on a well-defined proteome set, eg; one or a few species. What's your input? metagenomic sample or such?

I see no use in doing a taxonomic classification , moreover, as you already are suspecting interpro is not designed to do that. It is a collection of (curated) databases that contain conserved protein domains. As such those domains are present in a huge number of species known.

[drikaul]

Thanks for your response! Yes, it's a metaproteome sample, and I'm trying to run both blastp and interproscan to see if there is a difference in the results. But I'll take your advice and stick to protein domain annotations from interproscan, instead of trying to retrieve taxonomic info.

[lieven.sterck]

good to hear.

blastp and interproscan have very different goals and approaches though. True they both to some extent work on similar principles, but I see no clear gain in comparing their results to each other.

Since you're doing a blastp already anyway, analysing that output will give you a much better view of the taxonomic classification of your input proteins (eg. you can for each best hit ask the taxonomic classification and project that on your input sequence. Even better would be to use a more sophisticated approach).