How to get non-resistance genes from PFAM families

Question

[rpanat57]

Hello,

This is Rahul Panat, and my major is statistics. I am new in the area of Bioinformatics, and working on data analysis of resistance genes/non-resistance genes. Recently, I came to know that it is possible to obtain non-resistance genes from resistance gene (R-gene) using PFAM families and the quote that I collected from the internet is below.

"The negative sample was acquired from PFAM families due to the intimate relationship between R-gene and its protein sequence. No-duplicates PFAM of R-gene were removed from the whole PFAM families database. We got negative families here, and the longest sequence of proteins was fetched in each negative families."

As I am not clear about this quote, I have one query regarding collection of non-resistance genes. Can you please reply me (with step-by-step procedure) how we can collect non-resistance genes from PFAM (http://pfam.xfam.org/) using resistance genes?

I will appreciate much if you kindly help me in this regard.

Thank you.

Sincerely, Rahul

[Jean-Karim Heriche]

This is a question not a post about bioinformatics tools so I switched the type accordingly.

[rpanat57]

Thank you for the correction, Heriche.

Answer 1

[Philipp Bayer]

I assume that they made a list of PFam domains involved in resistance (NBS-LRR etc.), then took all PFam domains that are not in that list.

The paper you mentioned has no Supplementary Materials where you'd normally find these names but luckily the authors put up their negative set of non-resistance genes here, then you don't need to reinvent the wheel for them: http://server.malab.cn/rg_lc/Help.jsp

[rpanat57]

Thank you very much, Philipp. I know that the authors put up their negative set of non-resistance genes at http://server.malab.cn/rg_lc/Help.jsp. However, I am wondering how they found these non-resistance genes from Pfam using resistance gene set (which is also available at the same link)! I am not getting the logic or steps behind this. Couple of days ago, I tried to reach the authors via email but didn't get any reply from them yet.

[Philipp Bayer]

I'm sure that if you run Interproscan with the Pfam database on their set of resistance genes they'll all have the usual R-gene domains, copied from this paper:

CC
CC-NBS
CC-NBS-LRR
CC-TIR-NBS
LRR
NBS
NBS-LRR
TIR
TIR-NBS

Then you label all the remaining genes as non-resistance.

[rpanat57]

Thank you. I'll try it.