[phenixbb] Remodelling using Phenix

Fri May 4 01:55:35 PDT 2007

Ethayathulla,

from your description, it seems likely that the last 10 residues of the C-termus may just be disordered, esp if you observe this for each monomer.  if that is true, no autobuilding program will help you out (as far as I know), since they aren't as efficient building into density that is very weak or not present at all.  I think manually building may be your best option which doesn't sound too bad since it is only 10 residues/monomer.  If you aren't using Coot for building, i HIGHLY recommend it, since it makes these types of tasks very trivial.  With that, there are a few other options you might try, but I recommend trying to manually build as well, since these may not work.  

1- Have you tried Resolve's Prime and Switch?  While I have only used it once, there are many examples of PnS significantly helping bring out some otherwise weak/missing density.  You may run this from a SOLVE/RESOLVE installation via command line or, [I think] from within the PHENIX GUI (if i recall correctly).  here is a link for more info...

http://www.solve.lanl.gov/Resolve/html_resolve_manual/resolve_table_of_contents.htm

http://www.solve.lanl.gov/Resolve/html_resolve_manual/resolve_introduction.htm#ps

http://www.solve.lanl.gov/Resolve/html_resolve_manual/resolve_examples.htm

2- Another option might be to either do a round of density modification and/or calculate a full-omit map.  People have reported either of these can sometimes improve weak density.  Again, i think these are options that you can explore within PHENIX.  It might be worth your while to try them before and after you manually build the 10 extra residues/monomer....even if you only build the Calpha trace or polyA.  Check out the manuals for info on how to set some of these things up if you aren't familiar, or just post your questions here again, someone will surely help.  

3- and if you must venture away from PHENIX (kidding), then the folks at GlobalPhasing might recommend you give Buster-TNT a try.  While i have never used it, I have heard Bricogne speak wonders about the density maps that B-TNT produces, giving examples of much improved density maps, esp where weak/no density was previously observed.  Their people are awesome (but not as awesome as the PHENIX support, hehe) about helping people out with questions and with getting the software up and running so it can be used.  

Again, if it was me, I would first manually build in the 10 residues from the density present, even if just the Calpha or polyA.  Then do density modification and make maps to see if any additional density is present for the side-chains.  Even a sigma A-weighted 2fo-fc map should help you decide if the build was useful or not.  If there is no improvement, then it is very likely that this region is simply disordered and the programs probably won't help (in my little experience).  It is quite common for the C- and/or N-termini of proteins to be disordered, esp if they are recombinantly expressed and have an engineered 'linker' for tag removal.  If things do look better after DM, then you can probably build as best as you can, and allow the b-factors to inflate some (or however you deal with this).  

I hope some of this helps you out and you are able to build in your last 10 residues/monomer.  If any of the programs mentioned above works out for you, let us know, it would be good information for a rainy day.  Sorry I couldn't have been more help.  Best of Luck on your project!

Cheers,
Nick

  ----- Original Message ----- 
  From: Ethayathulla A.S 
  To: phenixbb at phenix-online.org 
  Sent: Thursday, April 19, 2007 2:18 AM
  Subject: [phenixbb] Remodelling using Phenix

  hi

  I am doing structure at 2.3A resolution. It is a dimer not NCS. I have probelm in tracing last 10 amino acids in the C-terminus in both chains. Both chains come close each other and winds up. So Iam finding difficulty in tracing the chain. I used phenix for rebuilding with all the amino acid included. But autorebuilding is not rebuilding properly, whatever input is given it rebuilds on that. Actually it should be a helix and only at lower cuff off we can observe density for main chain so I tried phenix rebuild to improve it but it's not working. Is there any option to use phenix. 

  I used rebuilding with option automatic and input coordinates and sequence.

  plz suggest do improve the phases.

  -- 
  A.S.Ethayathulla
  Department of BIophysics
  All India Institute of Medical Sciences
  Ansai Nagar
  New Delhi-110029
  India 

------------------------------------------------------------------------------

  _______________________________________________
  phenixbb mailing list
  phenixbb at phenix-online.org
  http://www.phenix-online.org/mailman/listinfo/phenixbb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20070504/053c05ac/attachment-0004.htm>