Dear Alex, Sorry yes I was focussing on the crescent density, since I imagined the metal atom assignment was secure. SInce Zn and Fe are now mentioned you would need to consider a tuned Xray wavelength anomalous experiment to be sure it is nickel, although the circumstatial evidence looks promising already.
Re the series termination effects; see eg:- A. Minichino, J. Habash, J. Raftery and J.R. Helliwell “The properties of (2Fo-Fc) and (Fo-Fc) electron-density maps at medium-to-high resolutions” (2003) Acta Cryst D59, 843-849. and references therein. That said my main idea of a cause of the crescent density being due to severe ice rings is not the case, as you now describe, so I am not sure what causes what you have there around your nickel ions. Greetings, John On Thu, Feb 24, 2011 at 5:02 PM, Alex Singer <[email protected]> wrote: > Hi -- thank you for all your help. The majority opinion seems to be a metal > for the sphere (Ni from the Ni-affinity column, which (Joe Patel, correct) > was used during purification, but Zn and Fe were also mentioned), and either > water molecules, bis-tris or some other small molecule forming the crescent. > Just looking at the density, the occupancy would seem to be quite high, so > I'm surprised that a Ni ion (or a contaminating metal ion) could have gone > through the purification and still remained at high enough concentration to > be clearly visible in the crystals. However, I'll still try this but first > some points of clarification and questions which you can either email me > seperately or post to the the group. > > a. it was collected at beamline 19-BM at Argonne, so radiation damage is an > issue. Thierry Fishmann -- for the gln residue, there was difference > density for the gamma carbon after the first conformation was modeled in, > thus the addition of the second conformation, which I agree is suspect. > What does the radiation damage do chemically and would that make the gamma > carbon more mobile? > > b. Jeffrey D Brodin -- how did you model in the bis-tris? Looking at the > bis-tris molecule from Hic-up, was the N at the centre of the crescent and > the O6 and O8 at the edges? > > c. JR Helliwell -- there are 4 molecules in the AU, but two H138's are > pointing into the solvent. Thus the molar ratio of protein molecules to > "thing 1" is 4:1. Also looking at the images, I see no ice rings -- the > images look pretty good. Can you tell me more about the series termination > effects? > > Again thank you for your help and I'll let the group know how it worked out. > > Alex > > -- > Dr. Alex Singer > C.H. Best Institute > 112 College St. Room 70 > University of Toronto > Toronto, Canada, M5G 1L6 > 416-978-4033 > -- Professor John R Helliwell DSc
