Hi Katherine,
One possibility could be that you have occupancy of 0.5 and that ligand
binding at one site in the crystal distorts a second site so it cannot bind
ligand.
Occupancy refinement is not always very stable.
If you are looking at two structures on top of one another - refinement needs
to be done carefully.
One thing I sometimes try is setting occupancy at various levels for the two
different conformations you have modelled, e.g. 0.2/0.8, 0.3/0.7, 0.4/0.6,
0.5/0.5, 0.6/0.4, 0.7/0.3, 0.8/0.2
Then check all maps carefully.
Sometimes soaking a crystal can change space-group and you may need to
reprocess data in lower symmetry cells to check. If you have only one molecule
in asymmetric unit might be worth trying the three P21 cells
(with beta=90 degrees).
[However, if you have to use DMF (rather than DMSO) to get your ligand into
solution - I would ask your chemists to make/give you some more soluble
analogues. ]
Ben Bax
Ben Bax
Reader in Structural Biology
Medicines Discovery Institute
Cardiff University
Main Building
Park Place
Cardiff
CF10 3AT
Email: b...@cardiff.ac.uk
Ben Bax
Darllenydd mewn Bioleg Strwythurol
Sefydliad Darganfod Meddyginiaethau
Prifysgol Caerdydd
Prif Adeilad
Plas y Parc
Caerdydd
CF10 3AT
Ebost: b...@cardiff.ac.uk
On 20 Dec 2019, at 04:57, Katherine Lim <katherine....@research.uwa.edu.au>
wrote:
Hi all,
I apologise in advance for the long post. I am working on solving a structure
that looks like it could have a ligand bound in the active site. My data was
obtained from a crystal of just the soluble domain of my protein that had been
soaked overnight in the ligand solution. The apo crystal structure is already
known and so I have solved my structure using phaser MR. I have attached the
Aimless report output of my structure at the end of this email. The current R
values I have after refinement and adding in all the waters are Rfree: 0.2413
and Rwork: 0.1897. I can clearly see green density that is much larger (only
disappears when I contour the Fo-Fc map to about 6 A) than in my control
crystal that had been soaked in the same concentration of just solvent (I had
used DMF).
I am struggling to add in my ligand as it doesn't seem like the entire ligand
can fit in the green density. We think it may be because we have only used the
soluble domain and so the ligand isn't held very securely since the
transmembrane domain is missing. I have been trying to fit in smaller sections
of it and doing an occupancy refinement. So far I have been able to get part of
the ligand in with an occupancy of about 0.6 but after the refinement run,
phenix.refine seems to move this part of the ligand slightly out of the area
where I had tried to fit it into the green density. Interestingly, there isn't
a big red density in the area that this ligand section has moved to. I would
appreciate any advice on how I should proceed with trying to figure out if I
have tried the correct section of the ligand and the kind of refinement
settings to use with a weak binder.
Space Group P212121
Unit cell abc 84.76, 89.84, 91.55
unit cell alpha beta gamma 90, 90, 90
OVERALL LOW RES HIGH RES
Low res limit 45.78 45.78 1.94
High res limit 1.9 9.11 1.9
Rmerge 0.234 0.052 1.684
Rmeas 0.244 0.054 1.768
Rpim 0.069 0.016 0.527
Total # observations 663073 6282 36851
Total # unique 55174 579 3359
I/sigma 7.5 27.9 1.4
CC ½ 0.997 0.999 0.747
Completeness %
99 98.7 94.7
Multiplicity 12 10.8 11
Katherine Lim
PhD Candidate
School of Biomedical Sciences; School of Molecular Sciences
Marshall Centre for Infectious Disease Research and Training
The University of Western Australia
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
<https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1>
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
