Hi, There are a number of possibilities that come to mind:
1. Use the "ROTATE AROUND" option of the brute-force rotation function in Phaser to generate a set of rotation angles similar to your real-space solution, then use each of these for a full translation search. 2. Use my old program BEAST to do exactly the limited 6D brute-force search that you ask about, in one run. But this is probably not better or even faster than doing the full translation search with the fast methods in Phaser, which is why we didn't bother making this kind of search convenient in Phaser. 3. The map around your positioned model should look a bit more like the real structure than the model does, so cut out the density (using a mask constructed from your coordinates) and search for the second copy in Phaser using that density. This is similar to what we did to solve the angiotensinogen structure (supplementary material in Zhou et al, Nature 468, 108-111, 2010). Cutting out density can be a bit tricky, so we describe how to do that on our web page: http://www.phaser.cimr.cam.ac.uk/index.php/Using_Electron_Density_as_a_Model. The procedure is still a bit tricky after that, so we're working on automating it. You might want to ask for more advice if you pursue this option. 4. Carry out refinement on the one-molecule solution, then use the refined model to search for the second copy. I think this strategy is part of the Balbes pipeline, and we've had some good results with that kind of thing too. 5. Wait a bit, and then some new features in Phenix and Rosetta will be available so that you can use Rosetta to rebuild your one-molecule model, adding information from the density to the sophisticated energy functions and conformational search tools in Rosetta. I should mention that this is a project involving Tom Terwilliger, David Baker and a lot of work by Frank DiMaio in David Baker's group. Regards, Randy On 7 Dec 2010, at 21:38, Arnon Lavie wrote: > Hi there: > > The situation: We are facing difficult molecular replacement: we believe we > have two molecules in the ASU, but phaser/molrep find only one. Using the > electron density calculated using this single molecule, we have manually > placed the 2nd molecule, albeit not good enough for rigid body refinement. > > Our strategy: We are looking for a program to do a 3 dimensional search > around the current position of the 2nd molecule. Maybe one that calculates > R-factor at the different positions, to allow to identify the correct one. ... > > Does anyone know of such a program, or an alternative approach? > > Thanks. > > Arnon > > -- > *********************************************************** > Arnon Lavie, Professor > Dept. of Biochemistry and Molecular Genetics > University of Illinois at Chicago > 900 S. Ashland Ave. > Molecular Biology Research Building, Room 1108 (M/C 669) > Chicago, IL 60607 > U.S.A. > Tel: (312) 355-5029 > Fax: (312) 355-4535 > E-mail: la...@uic.edu > http://www.uic.edu/labs/lavie/ > *********************************************************** ------ Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: + 44 1223 336500 Wellcome Trust/MRC Building Fax: + 44 1223 336827 Hills Road E-mail: rj...@cam.ac.uk Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
