[phenixbb] Only one solution in AutoMR with very negative LLG

Ed Pozharski epozh001 at umaryland.edu
Tue Mar 15 07:36:11 PDT 2011

On Mon, 2011-03-14 at 18:32 -0400, Zhang yu wrote:
> I am working on a DNA-protein complex, the protein apo structure is
> already known. Recently I got a dataset of the DNA-protein complex, I
> tried to find a solution by autoMR in Phenix with the known protein
> coordinate as template. I got only one solution with very negative LLG
> (around -4000), and after rigid body refinement, the both Rwork and
> Rfree is around 0.55.  Was there anyone in the same situation as me?
> What does that mean if there is just one solution while with very high
> negative LLG?

Could mean several things, but one thing is for sure - R~55% suggests
that molecular replacement did not work.  One possibility is that your
protein undergoes conformational change when it binds to DNA.  If it has
distinct domains, you may be able to get a solution if running them as
separate models.

Negative LLG could mean that you did not correctly guess the unit cell
content.  Or maybe the space group is wrong.  You may want to post the
phaser log-file, since it's not obvious to me what you mean by only one
solution.  What were the Z-scores at rotation/translation steps?

"I'd jump in myself, if I weren't so good at whistling."
                               Julian, King of Lemurs

More information about the phenixbb mailing list