Hi Obayed, even though there is 20% sequence identity you may be able to get a very good homology model, especially if there is more than one protein structure in the PDB with 20% homology. Then you can overlap the pdbs and find out what structurally needs to be preserved as opposed to what is in total homology preserved. Typically it is the position of turns residues G, D, S, P, N etc. You won't know until you thread your protein through both pdbs and compare them all. Swiss Pro's Expasy has an easy program that will take an alignment with a pdb and generate a homology model with loops spliced in and energy minimized. There are many other more or less complicated programs, but it's a good one to start with. Paul
Dr. Paul Kraft Structural Biologist cell 586-596-2770 email: haresea...@yahoo.com email: kraft_proteome_resea...@yahoo.com This communication and any attachments contain information which is confidential and may also be privileged. It is for the exclusive use of the intended recipient(s). If you are not the intended recipient(s) please note that any form of disclosure, distribution, copying or use of this communication or the information in it or in any attachments is strictly prohibited and may be unlawful. If you have received this communication in error, please notify the sender and delete the email and destroy any copies of it. E-mail communications cannot be guaranteed to be secure or error free, as information could be intercepted, corrupted, amended, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept liability for any such matters or their consequences. Anyone who communicates with us by e-mail is taken to accept the risks in doing so. --- On Tue, 7/19/11, James Stroud <xtald...@gmail.com> wrote: From: James Stroud <xtald...@gmail.com> Subject: Re: [ccp4bb] Off Topic: How to delete loops from a protein To: CCP4BB@JISCMAIL.AC.UK Date: Tuesday, July 19, 2011, 2:37 AM I've found that predator is one of the best services of this sort: Ref: PROTEINS: Structure, Function, and Genetics 27:329–335 (1997) Server: http://mobyle.pasteur.fr/cgi-bin/portal.py?#forms::predator The server is slow but the service is good. James On Jul 18, 2011, at 10:47 PM, Francois Berenger wrote: > Hi Obayed, > > If I understood your question well, > you are looking for something called "secondary structure prediction". > > I googled these keywords and found this server: > http://bioinf.cs.ucl.ac.uk/psipred/ > > You may find other interesting servers on the web and > some literature comparing them. > > I think such methods need only the sequence of your > protein to predict its secondary structures. > > Hope this helps, > Francois. > > On 07/19/2011 02:14 PM, Eric Larson wrote: >> Hi Obayed, >> >> you could give in situ protolysis a try. This is where you add a bit of >> protease along with you target protein to the crystallization drop. It >> has been quite successful for the folks at the SGC. Here are the >> relevant references: >> >> Dong A, et al. In situ proteolysis for protein crystallization and >> structure determination. Nat Methods. 2007 Dec;4(12):1019-21.PMID: >> 17982461. (http://www.ncbi.nlm.nih.gov/pubmed/17982461) >> >> Wernimont A, Edwards A. In situ proteolysis to generate crystals for >> structure determination: an update. PLoS One. 2009;4(4):e5094. PMID: >> 19352432. (http://www.ncbi.nlm.nih.gov/pubmed/19352432) >> >> good luck, >> >> Eric >> >> ================================ >> Eric T. Larson, PhD >> Biomolecular Structure Center >> Department of Biochemistry >> Box 357742 >> University of Washington >> Seattle, WA 98195 >> >> email: larso...@u.washington.edu >> ================================ >> >> On Mon, 18 Jul 2011, Obayed Ullah wrote: >> >>> >>> Hi all >>> >>> I wrote last time but got only one feedback. I know some of you guys >>> must have this experience that how to delete loops from the >>> protein. Please help me with suggestions. >>> >>> I am working with a human protein which have around 20% sequence >>> identity with the other proteins of the same family. Structure >>> of some of the proteins from this family have been solved. All the >>> solved structures have around 20% identity with my protein. I >>> am trying to crystallize the protein but it looks like very hard to >>> get crystal. I have tried different N and C terminally >>> truncated constructs for crystallization but no crystal. My feeling is >>> that probably there is some flexible loops with in the >>> protein which limiting the crystallization. >>> >>> So I want to delete the loops with in the protein (not to truncate in >>> the terminal, I already have done this). I am not asking >>> suggestion about how to delete the loop rather how to decide where the >>> loop is. I am not sure how much it will be helpful to get a >>> homology model of such a protein having low sequence identity. Is >>> there any strategy to decide where the loop could be? Does >>> anybody know any established/ rational method to do that. >>> >>> Waiting for your suggestions >>> >>> Obayed Ullah >>> >>> >>> >>> >>>
