Denis, Others may have addressed this, but you have not mentioned your refinement conditions. At 2.4 Å, are you using what and how early in the refinement is your work? NCS restraints, group B refinement, individual B refinement, etc. A little more information would help the BB readers respond to your query. Are the overall Bs for each subunit similar? We have had cases where one monomer has relatively low Bs (30-40), while the other subunit(s) are noticeably higher (say in the 50s). If that is the case, you will not necessarily see the same detailed structure in each subunit.
However, Herman’s comment about doing superpositions is the first thing I would do, with one caveat: do not rely on just a global superposition (A onto B). Subtle changes can be masked. If your protein/enzyme has two domains, superimpose each domain (A1 onto B1 and A2 onto B2) then see what happens. You may see some subtle shifts. What Emily suggested is also correct, if you have an assay, but remember that resolving water can be hit or miss depending on the environment (chemical and crystallographic) in the asymmetric unit. One crystallographic phenomenon people tend to forget is that the noise across an asymmetric unit is not uniform (due to all sorts of accumulated phase error, truncation errors, etc.), so the metal site in one monomer may be in a slightly lower noise environment than the metal site in the second monomer. It may be there, but just not resolved. Bottom line is to be cautious and skeptical with any hypothesis you want to test about this water. Cheers, Michael **************************************************************** R. Michael Garavito, Ph.D. Professor of Biochemistry & Molecular Biology 603 Wilson Rd., Rm. 513 Michigan State University East Lansing, MI 48824-1319 Office: (517) 355-9724 Lab: (517) 353-9125 FAX: (517) 353-9334 Email: rmgarav...@gmail.com **************************************************************** > On Nov 28, 2017, at 2:04 PM, Denis Rousseau <denis.rouss...@einstein.yu.edu> > wrote: > > Hi BB members > > I have a homodimeric protein, which contains metal centers. In several > different derivatives I find a water molecule on a metal center in one > subunit which is absent on the other. It is always the same subunit that > contains the water molecule. The resulution is ~2.4 A. Is this an artifact or > a functional difference? If it is truly homodimeric I would expect > differences to be random. The space group is P212121. > > Thanks for the advice. > > Denis Rousseau
