Extracting Domains

Question

[Hari]

Hi all, I used CDD for extracting domains from my protein sequence,the results are little difficult to analyse i mean i have around 100-120 protein seq.of Rho Gtpases i need to extract the domains (seq with start and stop positions),well i tried to use web based tools to no avail, i just to get just the ph and dh domains from the sequence?Please suggest me some methods.

Thanks in advance

[Hari]

I really found pfam very useful for extracting domains.

Answer 1

[Mateusz]

You could also use batch-search on Pfam. As output you will get text file which could be easily parsed using python. I was using it to extract domain architectures of proteins. It does contain name of domain, start, stop etc. Unfortunately it doesn't contain sequence, but BioPython have libraries to download sequence so if you have start and stop positions it shouldn't be a problem to retrieve domain sequence.

Perhaps there is similar solution for CDD.

[Michael Schubert]

CDD does support batch search online: http://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi

[Hari]

thank you for ur reply,I am still waiting for the result and hopefully i can do some further analysis soon

Answer 2

[Michael Schubert]

Take a look at this:

http://www2.warwick.ac.uk/fac/sci/moac/students/peter_cock/python/rpsblast/

Ignore the BioPython part if you wish. rpsblast does give you the match boundaries as a result, even in tab-delimited format.

If you get stuck anywhere post a comment.

[Hari]

Hi i am yet to try this method will let you know once i tray..thanks for the help

[Hari]

Hi I did not try this method may be i will do it later as I was able to do it with uniprot search and also CDD batch search was helpful thank you..

Answer 3

[Sean Davis]

I would suggest looking at a biomart interface to your data source. In particular, you could try uniprot:

http://www.ebi.ac.uk/uniprot/biomart/martview/

[Hari]

Thank you for your reply,well i am not really getting right results it does not exactly give me just the domains(i mean the exact sequence) but still i am trying to refine my search....

[Hari]

Biomart was also helpful but pfam results were much better ,i was able to directly download the domain sequence with start and end postions