Well, I think it's more of a data-vs-geometry battle, with the geometry dragging in the side chains towards the calciums, and the data pulling them out. When I real-space refine them in Coot, some of the Ca-O distances are pretty different from the "ideal" values. Maybe I could just exclude them from xyz refinement in Phenix after making sure they're correct in Coot? I'll give that a shot. Kol tuv, Jacob -----Original Message----- From: Boaz Shaanan [mailto:[email protected]] Sent: Tuesday, December 29, 2015 7:52 AM To: Keller, Jacob; [email protected] Subject: RE: Modifications for Ca-binding site Hi Jacob, Sucking-up neighboring residues into the Calcium (or other heavier atoms) density is a known issue. Turning-off the Ca-O or other distance restraints around the Calcium is the last thing you want to do, in my experience. You'd better check those distance restraints and verify that they're compatible with the distances in other (preferably high resolution, of course) structures with calcium binding sites similar to yours and then make sure they're given the appropriate weight during refinement. That's my view but Pavel will give you more advices I'm sure. Cheers, Boaz Boaz Shaanan, Ph.D. Dept. of Life Sciences Ben-Gurion University of the Negev Beer-Sheva 84105 Israel E-mail: [email protected] Phone: 972-8-647-2220 Skype: boaz.shaanan Fax: 972-8-647-2992 or 972-8-646-1710 ________________________________________ From: [email protected] [[email protected]] on behalf of Keller, Jacob [[email protected]] Sent: Tuesday, December 29, 2015 4:44 AM To: [email protected] Subject: [phenixbb] Modifications for Ca-binding site I am getting a lot of problems with my calcium binding sites--there is clear density for the calciums and the protein side chains, but Phenix keeps sucking the side chains out of their density towards the calciums and also the calciums show some positive difference density, perhaps due to too-high b-factors (or potentially partial occupancy by Na?). I think what might work best is to turn off the Ca-O bond length restraints somehow, but don't see where/how to do that. I tried releasing them from geometrical restraints, but that made things worse, since things went haywire in the sidechains themselves. Is it possible to make Phenix ignore the Ca-O bond distance restraints? JPK ******************************************* Jacob Pearson Keller, PhD Looger Lab/HHMI Janelia Research Campus 19700 Helix Dr, Ashburn, VA 20147 email: [email protected] ******************************************* _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]