[phenixbb] automating setting up parallel refinement jobs

[Francis E Reyes]

On Feb 17, 2011, at 5:21 PM, Donnie Berkholz wrote:

> Bjørn P. Pedersen, J. Preben Morth and Poul Nissen. "Structure
> determination using poorly diffracting membrane-protein crystals: the
> H+-ATPase and Na+,K+-ATPase case history" Acta Cryst D66: 309-313
> (2010).


I've tried this out with a few of my own structures. In good cases (>  
50% coverage of the ASU with your search model), there is a  
correlation between high TFZ's and high anomalous peaks (i.e. if you  
took their Figure 1, and then made the same thing but instead scored  
for high anomalous peaks). However,  I currently have a similar case  
to their MHp1 example (30% coverage of the ASU of a fold that I  
believe to be in my structure) and using their pipeline,  my TFZ's are  
in the 7's with very low anomalous peaks... i.e. no solution.

Another interesting case is if you have anomalous data and an MR  
model, could anomalous peaks be a filter for scoring MR solutions  
(have a separate column in phaser next to TFZ that checks the  
anomalous map with the MR model phases). I had a conversation with  
Randy Read about this before and in his test cases, it didn't seem  
reliable.

I wonder if an anom LLG map would be better at teasing this one out.  
Or even better yet, if you have a bound cofactor (my case) or know  
exactly where your heavy atoms are (Se-Met), it would be interesting  
to only check the anomalous peaks in a given region (basically search  
the anomalous map given a pdb selection of the model) .

Suffice to say, I never do a single phaser run anymore, but  
parameterize RMSD and reso. It's a lot of jobs, but when it's passed  
to the cluster, it goes relatively quickly.


F
>

---------------------------------------------
Francis E. Reyes M.Sc.
215 UCB
University of Colorado at Boulder

gpg --keyserver pgp.mit.edu --recv-keys 67BA8D5D

8AE2 F2F4 90F7 9640 28BC  686F 78FD 6669 67BA 8D5D