[phenixbb] assign multiple ligand conformations
tg at shelx.uni-ac.gwdg.de
Tue Feb 11 00:55:40 PST 2014
I would split the entire ligand in coot twice, not only the head. You
can set the initial occupancy with coot, too.
This has the advantage that the three conformation gain greater
flexibility than what you might expect from the current density and, if
the putatively non-disordered tail does not refine stably, you have a
good indicator that you are overfitting and that e.g. your data are not
strong enough for occupancy refinement.
On 02/11/2014 05:35 AM, Wei Shi wrote:
> Hi everyone,
> I am working with protein-ligand complex structure. For the head group in
> one of the ligands in the structure, two more positions in addition to the
> current position look possible according to the blue density in Coot. I am
> thinking of assigning multiple occupancies/positions to this head group.
> So, I am wondering what I should do to achieve this. Is the following
> strategy right?
> 1. Fit the head group to the density at three different positions
> separately in Coot and save them separately.
> 2. Find out the atoms that are different among the three ligand positions
> and combine the same atoms with different positions and change the
> occupancy of those atoms to something between 0 to 1.
> 3. Use Phenix. refine to refine. The occupancy refinement should be
> automatically on, right?
> Is this the best way to assign multiple occupancies/positions to the head
> group? Thank you so much!
> Also, anything to look after refinement which can tell me whether I am
> assigning the right multiple conformations to the ligand?
> Thank you so much!
> phenixbb mailing list
> phenixbb at phenix-online.org
Dr Tim Gruene
Institut fuer anorganische Chemie
GPG Key ID = A46BEE1A
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 234 bytes
Desc: OpenPGP digital signature
More information about the phenixbb