You could try an LLG anomalous map from Phaser as well to clarify the matter: in my experience, even sulfurs are found, and your Rb should have f” ~0.75 even at 1 Ang xrays. Also, you could refine in Refmac including anomalous data, and refine occupancy of Rb.
Having both modeled into the data seems fine to me, and I think fairly common. Just set occupancies accordingly. Incidentally, it looks to me like a weird place for a cation to dwell, but if the data support it… JPK From: CCP4 bulletin board [mailto:[email protected]] On Behalf Of Jarrod Mousa Sent: Friday, April 24, 2015 9:44 AM To: [email protected] Subject: [ccp4bb] Lipid molecule Hi all, Picture attached. I solved the structure of a membrane protein using LCP. When trying to identify the cation-binding site, I co-crystallized with Rb+. I see strong density corresponding to Rb+, and it was confirmed in an isomorphous difference map. One problem: in the native structure I see strong density for a lipid molecule from the LCP, in which I can model in very well. In the Rb+ structure, I see more of a blob of density at low rmsd, but when at higher the spherical Rb begins to show. Based on the isomorphous differnece map showing nice spherical denstiy for Rb+, it seems the crystal I shot contains protein molecules with the lipid and others with Rb+. I currently have Rb+ modeled in, but there is lots of extra density surrounding it, corresponding to the lipid molecule. Should I put both molecules in the structure, even though it doesn't make much sense and would results in odd lipid-cation clashes? Or should I just leave the Rb+ without lipid since this makes sense chemistry-wise? Thanks, Jarrod ------------------------------------- Jarrod J. Mousa, Ph.D. Bruner Laboratory Department of Chemistry University of Florida email: [email protected]<mailto:[email protected]>
