Hi all, I am modeling some proteins but in my search for homologs, I have found that out of ~1500 residues, the available templates only cover about 600 residues. My question is, how do I model the part of the target that is not covered by the templates? I haven't done any modeling before and will really appreciate some answers.
Pauline.
•
written
by
paulawambui •
How do you find the homologs? Do you perform a single blast vs pdb database?
You should try to find remote homologs for the uncovred region of your protein. For example, HHpred: http://toolkit.tuebingen.mpg.de/hhpred
Also, you can try with the I-TASSER server: http://zhanglab.ccmb.med.umich.edu/I-TASSER
Good luck!
I used hhpred and the results showed that only a section of the target was covered by the templates returned i.e. in a 1502 protein, almost all the templates were falling in the region ~146 to ~746.
Many thanks.
If you haven't found any template and are sure about that, ab-initio modelling is your last chance.
Before that, you should try FFAS and automatic homology modelling servers to see what sequences are retrieved and what part of your protein can be modelled.
Also have you checked automated models databases such as PMP and Modbase ?
I have only used hhpred. I will try your suggestions.
Many thanks.
Log in to answer this question.
