Hi Chris, this certainly improved your model with respect to the data including all their errors. How do you know this did not make your model worse with respect to chemistry, respectively to what is inside your crystal?
Best, Tim On 05/08/2014 07:26 PM, Chris Fage wrote: > Hi Everyone, > > Thank you for the advice, especially Pavel's. My issue has been > resolved. I lowered the occupancy and B-factors of the metal ions in > the .pdb and ran phenix.refine for 10 cycles. This removed most of the > negative Fo-Fc density. Anisotropic refinement of the metal ion > B-factors was also effective, dropping Rwork and Rfree by ~1% each. To > do this, I edited the "Individual B-factor refinement" values as > follows. > Isotropic atoms: not (element Zn) > Anisotropic atoms: element Zn > > Best, > Chris > > > On 5/6/14, Chris Fage <cdf...@gmail.com> wrote: >> I have used CNS before, but not for this sort of refinement. I see in >> the bindividual.inp file that I can "select atoms to be included"--it >> is defaulted at "known and not hydrogen". Do you know the proper >> nomenclature for selecting a Zn ion in chains A and B? >> >> Thanks, >> Chris >> >> On 5/6/14, Steven Herron <sherron_...@yahoo.com> wrote: >>> >>> Refining the occupancy will help your R-factor and flatten your density, >>> but you need to be careful to also refine the B-factor of the metal >>> ion. Don't refine both the occupancy and the B-factor during the same >>> run (the two are correlated at this resolution), refine the occupancy of >>> just the metal ion and then refine the B-factor of just the metal ion >>> (repeat as needed). I used X-plor/CNS to do my refinements, so it was >>> easy to refine the occupancy (or B-factor) of just the metal ion. After >>> a few rounds of refinement both parameters will stop changing and you >>> will have your answer. The final B-factor of the metal ion should be >>> similar to the amino acid residues that are coordinated to it. >>> >>> Soaking in several different ion concentrations and collecting >>> additional datasets is also a good idea (if you have the time). I did >>> this type of experiment once before (see: JBC 278(14):12271-7. [ Apr 4, >>> 2003]) (or: http://www.ncbi.nlm.nih.gov/pubmed/12540845). I soaked in >>> several different Ca2+ ion concentrations and was able to determine the >>> binding affinity for that calcium ion using crystallography. >>> >>> To make sure I was not stuck in a local minima, I would modify either >>> the occupancy of the B-factor of the metal while keeping the other fixed >>> and do a refinement. I even tried both large and small changes (both >>> increases and decreases in value). It always came back to the earlier >>> answer. >>> >>> Different Ca2+ ion concentrations can give some additional insight into >>> the metal binding site. Between the no-Ca2+ structure and the high-Ca2+ >>> structure there was a conserved Asp-residue that changed conformation. >>> So, I soaked in the appropriate amount of Ca2+ to see the residue in >>> both positions. There was a high correlation between the asp residue >>> orientation and the Ca2+ ion occupancy. >>> >>> Steven Herron >>> sherron_...@yahoo.com >>> >>> >>> >>> >>> On 5/6/2014 11:02 AM, Chris Fage wrote: >>>> Hi Everyone, >>>> >>>> In my 2.5-angstrom structure, there is negative Fo-Fc density >>>> surrounding a metal ion after refining in Phenix. From anomalous >>>> diffraction I am certain of the metal's identity and position in each >>>> monomer. Also, the ion is appropriately coordinated by nearby side >>>> chains. Should I be refining the occupancy of the ion in attempt to >>>> "flatten" the negative density? I am considering soaking the metal ion >>>> into crystals or cocrystallizing and collecting additional datasets. >>>> >>>> Thanks for your help! >>>> >>>> Regards, >>>> Chris >>> >>> >> > -- Dr Tim Gruene Institut fuer anorganische Chemie Tammannstr. 4 D-37077 Goettingen GPG Key ID = A46BEE1A
signature.asc
Description: OpenPGP digital signature
