Hi, Most of the answers are interpreting your question as follows, that you have a single Se-SAD data set that goes to 3.3A but only has anomalous signal to 3.8A. Is that correct? If so, then there are various ways to combine the MR and SAD phase information. As you might imagine, we like the procedure in Phaser. To expand a bit on what Nat said, Phaser takes the MR solution and treats it like a substructure (albeit one that has no anomalous scattering), then it uses log-likelihood-gradient (LLG) maps to find where anomalous scatterers should be added to improve the agreement with the data. In our experience, including the MR model in this way can give you a more complete anomalous substructure (including minor sites from statistical disorder of the Se-Met side chains), which should improve the phases somewhat. Because the anomalous scatterer sites and the MR model are treated as part of the same structure, and refined simultaneously, the phase information is combined naturally, without having to export part of the information through Hendrickson-Lattman coefficients. Refmac has many powerful features but, as far as I know, SAD LLG maps are not among them, so it's worth trying this even if you subsequently move to refinement in Refmac.
The other way your question could be interpreted is to say that you have a native data set to 3.3A and a Se-SAD set to 3.8A. If that's true, then the best approach would depend on how isomorphous the two data sets are. If they're isomorphous, then you should treat it as SIRAS, but if they're not then you might want to do some cross-crystal averaging. Again, in this scenario there are various ways of combining the experimental and model phase information. However, completing the other 2/3 of the structure when you only have experimental phase information to 3.8A is a daunting challenge, if you don't have NCS or something else in your favour. In your situation I'd still want to be working in the background on getting higher resolution native or SAD data, other crystal forms and/or other derivatives. Regards, Randy Read On 13 Jul 2011, at 02:08, Jiamu Du wrote: > Dear All, > I am now working on a low resolution phase determination (around 3.3 A with > Se anomalous signal around 3.8 A). > I can find the Se site and get the phase, but the density map is not so good. > Some part of the protein (about 1/3) has a homologue model which is also can > be found using Phaser. The homologue region has a good map while other region > only show a poor map. > I think the combination of experimental phase and MR phase might improve the > map. Is there anybody can help find which program can work on this? > > Thanks. > -- > Jiamu Du, Ph.D. > Postdoctoral Research Fellow > Laboratory of Structural Biology > Memorial Sloan-Kettering Cancer Center > RRL 269, 430 E 67th Street > New York, NY, 10021 > E-mail: d...@mskcc.org > Tel: (217) - 417 - 9897 > ------ 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
