[phenixbb] Best first map for RESOLVE after MR

Randy J. Read rjr27 at cam.ac.uk
Mon Aug 31 12:51:32 PDT 2009


A map from phenix.refine is probably better, if you're going to choose one, 
for two reasons:

1. As you say, it includes a proper bulk solvent contribution. 2. Phaser 
doesn't (at least yet -- we might get around to putting it in some day) 
refine the Sigma(A) values, so the weights are determined from the guesses 
about the RMS error of the model.

I think Phaser usually does a pretty good job of rigid-body refinement, but 
it shouldn't hurt to repeat this step in phenix.refine. If there are any 
domains that might move, this would be a good time to refine them as 
separate rigid bodies.


Randy Read
On Aug 30 2009, Pavel Afonine wrote:

>Hi Peter,
>I guess this my reply will not replace Tom's or Randy's one...
>> Should it be the 2mFo-DFc map directly from Phaser or would it be 
>> better to do rigid body refinement in phenix.refine and use the bulk 
>> solvent corrected 2mFo-DFc map ?
>- why don't you try using both options and choose the best outcome?
>- running a round of rigid body refinement in phenix.refine and using 
>its 2mFo-DFc map will not hurt at least or may make your model and 
>therefore the map better, so you will have a better starting point.
>phenixbb mailing list
>phenixbb at phenix-online.org

More information about the phenixbb mailing list