On 17 May 2012 21:27, Eleanor Dodson <eleanor.dod...@york.ac.uk> wrote:
> I am not familiar with CNS restraints, but whatever the program restraints > are - if you do a omit map, or just set the occupancies of the metal and > its surroundings to 0.00 and do a few cycles of refinement, I believe any > model bias will disappear and what you see will be pretty accurate > description of the metal. > When you include that atom in the refinement it is usually necessary to > at least set distance restraints up - otherwise the heavier metal density > will tend to swallow any water molecules around about > Eleanor > > > On 14 May 2012 15:19, Ed Pozharski <epozh...@umaryland.edu> wrote: > >> On Sat, 2012-05-12 at 08:48 -0400, Dave Roberts wrote: >> > However, I just want to make sure the metal environment is not due to >> > the fact that I did something wrong in my refinement script - thus >> > making it tetrahedral because it was refined as tetrahedral. >> ... >> > I don't use CCP4 for refinement, I use CNS, but I'm hoping somebody >> > will be able to guide me here. >> > >> >> AFAIK, CNS does not generate any geometry restraints for metal ions >> *automatically*. You may want to make sure (if you didn't yet) that you >> use a correct residue name, e.g. NI2 (and atom name NI+2) for Ni++. CNS >> libraries would use different values for the radii and scattering >> factors of these. >> >> >> -- >> "I'd jump in myself, if I weren't so good at whistling." >> Julian, King of Lemurs >> > >
