 |
 |
Hi Roger, I think your ideas are sound, but I would add some "prime-and-switch" density modification in resolve plus/minus ncs to try and improve the maps and cut down on phase bias. Hth, Dave -- Hand delivered by Androids On 1 Sep 2010 16:12, "Roger Rowlett" <[log in to unmask]> wrote: I am trying to find a MR solution for a large unit cell (R3:H, 158x158x196) with a relatively poor, but I think workable search model (30% identity, 50% similarity). The data set is decent to 2.4 A. I might be able to get better if necessary. I submitted the sequence of the target to the Phyre server, and it returned a PDB file derived from what I had already identified as the most likely successful search model. Phaser finds a reasonable solution with four dimers (Z=9-15 depending on the data set) for which the unit cell packing looks good. Phaser even assembles what looks like "correct" biological tetramers in the ASU and across the symmetry interfaces. The main chain appears to be mostly traceable, but the side chains are not all well-resolved, and I suspect from the better-defined regions, that the chain is misregistered by a residue or two throughout most of the structure. Assuming that an MR solution is possible, what is a good approach from here? FWIW, Phaser does not correctly place a poly-Ala/Gly model, although it may place three chains similarly to the MR solution I have with the Phyre model. My first thought is to: 1. Chainsaw the currrent solution, and attempt to identify and build in the correct register of the side chains. after refinement. 2. Do a low resolution refinement of the poly-Ala/Gly model to better thread the main chain? 3. Try EPMR with the Phyre model or the poly-Ala/Gly model 4. Give up and get real phases (but I'm so close now!?) Thanks in advance. -- ------------------------------ Roger S. Rowlett Professor Department of Chemistry Colgate University 13 Oak Drive Hamilton, NY 13346 tel: (315)-228-7245 ofc: (315)-228-7395 fax: (315)-228-7935 email: [log in to unmask]