Hi, You can also use Phaser log-likelihood-gradient completion, starting from your protein model, to find the Br atom(s). This is usually more sensitive than the traditional model-phased anomalous difference Fourier. This can be run either from Phaser-EP or from AutoSol. In Phaser-EP choose the "MR-SAD phasing" mode first. In either case, you have to choose the data file, a sequence file, and the molecular replacement model. In AutoSol, make sure that the GUI has identified the PDB file as a "Partial model". Good luck! Randy Read On 8 Feb 2011, at 21:11, Christian Roth wrote:

Hi Jason,

I am not an expert, but in my opinion it is enough to collect 1 dataset at the peak wavelength for bromide. Should be anomalous complete for sure. Process it without merging Friedel pairs. Solve the structure via molecular replacement with your known protein structure. Next step ist to create a map using the anomalous differences in your mtz with the phases from your mtz after refinement. Now you should see a peak in the density for your Bromine, if ligand is there. Than you have the position of your ligand There should be hopfully density from you normal maps. Put your ligand in it. The bromine dictates the orientation. Put this with the generated library file(phenix.elbow) in phenix and refine it.

I think this should work.

Good Luck and all the best.

Christian

P.S. Despite the fact that you did cocrystallisation, the ligand might not be there, or even at a position which is not the right one. Unfortunately I had already such a case.

Am Dienstag 08 Februar 2011 19:14:42 schrieb Jason:

...

Hello everyone,

I have a few crystals to be x-rayed next week. Before that I hope to get a clear idea about what I am doing (I am new to anomalous scattering).

Facts:

1. The crystal is a protein co-crystallized with a ligand 2. The protein structure is known. 3. The ligand has a heavy atom bromide (absorption K-egde=13.47Kev) 4. Data resolution is ~3 angstrom

Goals:

1. Locate the bromide position 2. Locate and refine the ligand

Questions:

1. Do I need to carry out MAD experiment at 3 wavelengths or there is some other easier way since the protein structure is known (I am not expecting big change of the protein structure itself)? 2. Assuming I have the MAD data, what should I do next using phenix to achieve the two goals listed above? Here are some thoughts: - Using phenix.hyss to locate the anomalous scatterers - Using phenix.autobuild to build the protein model (which data set to use?) - Using coot to add ligands to the protein structure (Is the relative position between the protein and the ligand known based on phenix.hyss and phenix.autobuild?) - Using phenix.refine to refine the ligand+protein complex

Thank you all for reading this.

====================== Jason Structural Biology Department University of Pittsburgh ======================

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

------ 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: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk