Dear Daniel, There are two questions here. First, is there likely to be enough signal to get a clear solution for the molecular replacement; second, will getting a solution for the molecular replacement be of any use to you?
On the first point, we have some new ways of estimating the signal of a molecular replacement search in Phaser. These depend on the completeness of the model, the estimated error of the model (relative to the resolution of the data) and the number of reflections. With 100% identity, your model should be almost perfect at 7A resolution, which is in your favour, but 6% of the a.u. is very much against you. In any case, the calculations indicate that there would be a marginal signal in this case (predicting an LLG of about 16 for a correct solution). If you were lucky and the part of the structure you can model were the best-ordered part of the crystal, then you could do somewhat better and may even get a clear solution, which usually requires an LLG of about 50 or more. In contrast, molecular replacement calculations with ribosomes can give very clear solutions at resolutions below 10A, because there are so many reflections in a large problem and the models can be reasonably complete. So it's very case dependent. On the second point, it's hard to imagine what you would learn from this. There are no methods to complete a very incomplete model at such low resolution, and you're not modeling a complex so you wouldn't learn about interactions. However, if you had some kind of heavy atom derivative or anomalous scatterer bound, then there might just be enough information from this to help to find the sites, and thus bootstrap the structure solution. Best wishes, Randy Read On 6 Nov 2012, at 11:06, Panne Daniel wrote: > Hi all, > > I was wondering if there are examples of successful MR with low resolution > data (~7A). Our protein crystallized in the P21 (a= 111; b= 211, c=126; > beta=99.7). We only have a partial search model (267 amino acids) that > accounts for ~6% of the AU (however at 100% identity). > > Has anybody been successful in such situations? > > Thanks, > Daniel ------ 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
