Re: [ccp4bb] tricoordinated ion?

[Roger Rowlett]

Before jumping to any conclusions, you should definitely see what this density looks like without the water molecules in this portion of the model. I think it would be difficult to decide on what the actual "bond lengths" are without removing model bias. If you model with waters, of course the electron density will conform to some degree to what is modeled there. You may find that without the waters, the electron density shrinks up a bit, or can be adequately modeled with a trigonal planar anion. On the other hand, maybe it is just a cluster of water molecules, or an anion at less than 100% occupancy. From your water model, however, it looks like you may have enough density for a full occupancy anion. FYI, the surrounding ligands, Arg + Glu + another H-bonding donor are certainly compatible with the few known bicarbonate binding sites in proteins. Roger Rowlett Klaus Piontek wrote: Hi Sebastiano, I had similar experiences with various crystal structures. After a lot of modeling (albeit at about 1 A resolution) I came to the conclusion that these atoms correspond to disordered solvent molecules, e.g. water and/or Zn. In other words the atoms have occupancies of < 1, but in the sum it should be 1. For CO2 or similar molecules the bond length are far too long (C-O is about 1.4 A or so). Good luck. Klaus Piontek Sebastiano Pasqualato wrote: Hi all, I wanted to ask you what would you model in the density in which I have at the moment modelled 4 water molecules, which are however too close to be waters, I guess (see attached image). My crystallisation conditions contain NaCl, MgCl2, Peg400, TrisHCl, TCEP, glycerol. I can't think at a tricoordinated ion like that... thanks in advance for the hints, ciao S -- Dr. Klaus Piontek Albert-Ludwigs-University Freiburg Institute of Organic Chemistry and Biochemistry, Room 401 H Albertstrasse 21 D-79104 Freiburg Germany Phone: ++49-761-203-6036 Fax: ++49-761-203-8714 Email: [EMAIL PROTECTED] Web: http://www.chemie.uni-freiburg.de/orgbio/w3platt/