Re: [ccp4bb] how to combine two derivative datasets that are not isomorphous?
How non-isomorhous is non-isomorphous? Can you give the cells and putative Spacegroups for A a B b? You can often get information at low resolution for this sort of problem if there is some level of similarity. Eleanor On 18 November 2014 21:15, David Schuller schul...@cornell.edu wrote: You don't mention any quality indicators on your derivatives, nor resolution. Presuming they actually have some decent phasing power, you may be able to generate phases maps using SIR phasing + solvent flattening. If you can do that for each isomorphous set, then you could combine them using multiple space group noncrystallographic averaging. Or, if the derivatives contain atoms with anomalous scattering, you could solve a derivative alone with SAD methods, or a derivative + native set with SIRAS. Once you get crude maps, you can once again try to combine the two sets with multiple space group averaging. As for MR, there are lots of things you can try. Once again, you have provided no detail on what the sequence similarity is, or any other factor that would allow us to judge the likelihood of success. You could search again with the model clipped down to poly-Ala. You could search again with any external loops trimmed off. If it is a multiple domain molecule, you could search individually with single domains. You could find additional search models, and search with a suite rather than single model. MR programs will suck up as much free time as you can provide them. On 11/18/14 15:01, joy yang wrote: Hi All, I have two derivative datasets (heavy atom A and B) and two native datasets (a and b), A and a are isomorphous, B and b are isomorphous, however, a and b (or A and B) are not isomorphous. I was able to make two difference patterson maps (FA-Fa and FB-Fb) and search for heavy atoms against them, the possible positions of heavy atom A and B are very close to each other in the unit cell (which seems to me that I am very close to a right phase, though not there yet), I am wondering if there is any means for me to combine the information from FA-Fa and FB-Fb as the two native datasets are not isomorphous? And also, I have a homology model which I tried molecular replacement and failed, is there any means for me to combine the information from the model too? Best, Bei -- === All Things Serve the Beam === David J. Schuller modern man in a post-modern world MacCHESS, Cornell University schul...@cornell.edu
[ccp4bb] how to combine two derivative datasets that are not isomorphous?
Hi All, I have two derivative datasets (heavy atom A and B) and two native datasets (a and b), A and a are isomorphous, B and b are isomorphous, however, a and b (or A and B) are not isomorphous. I was able to make two difference patterson maps (FA-Fa and FB-Fb) and search for heavy atoms against them, the possible positions of heavy atom A and B are very close to each other in the unit cell (which seems to me that I am very close to a right phase, though not there yet), I am wondering if there is any means for me to combine the information from FA-Fa and FB-Fb as the two native datasets are not isomorphous? And also, I have a homology model which I tried molecular replacement and failed, is there any means for me to combine the information from the model too? Best, Bei
Re: [ccp4bb] how to combine two derivative datasets that are not isomorphous?
You don't mention any quality indicators on your derivatives, nor resolution. Presuming they actually have some decent phasing power, you may be able to generate phases maps using SIR phasing + solvent flattening. If you can do that for each isomorphous set, then you could combine them using multiple space group noncrystallographic averaging. Or, if the derivatives contain atoms with anomalous scattering, you could solve a derivative alone with SAD methods, or a derivative + native set with SIRAS. Once you get crude maps, you can once again try to combine the two sets with multiple space group averaging. As for MR, there are lots of things you can try. Once again, you have provided no detail on what the sequence similarity is, or any other factor that would allow us to judge the likelihood of success. You could search again with the model clipped down to poly-Ala. You could search again with any external loops trimmed off. If it is a multiple domain molecule, you could search individually with single domains. You could find additional search models, and search with a suite rather than single model. MR programs will suck up as much free time as you can provide them. On 11/18/14 15:01, joy yang wrote: Hi All, I have two derivative datasets (heavy atom A and B) and two native datasets (a and b), A and a are isomorphous, B and b are isomorphous, however, a and b (or A and B) are not isomorphous. I was able to make two difference patterson maps (FA-Fa and FB-Fb) and search for heavy atoms against them, the possible positions of heavy atom A and B are very close to each other in the unit cell (which seems to me that I am very close to a right phase, though not there yet), I am wondering if there is any means for me to combine the information from FA-Fa and FB-Fb as the two native datasets are not isomorphous? And also, I have a homology model which I tried molecular replacement and failed, is there any means for me to combine the information from the model too? Best, Bei -- === All Things Serve the Beam === David J. Schuller modern man in a post-modern world MacCHESS, Cornell University schul...@cornell.edu
