 |
 |
Dear Nick, As mentioned a while back on the BB, the wwPDB X-ray Validation Task Force is now discussing updates to the validation tests, and one of the things we're looking at closely is whether there's something better than the RSRZ. That was the best we could come up with 5 or so years ago, but there could well be better alternatives. To a great extent, the properties of your crystal will limit how good the RSRZ can be for a particular residue. If part of the structure is poorly ordered, it's not your fault and there's nothing you can do about it! I've worked a fair amount on serpins, which tend to have some very poorly ordered parts (probably because conformational change is an essential part of their mechanism of action), so they tend to do badly on RSRZ no matter what you do. As some people have pointed out to us, an unintended consequence of judging structures by the mean RSRZ or the number of RSRZ outliers is that it creates a temptation to simply omit the less well-ordered parts of your model, even if you have a pretty good idea of where they are. That's not really what we want as a community, so that's why we need better validation metrics, or at least better ways of presenting them. The bottom line is that you should view the RSRZ as a pointer to where you should look for possible problems in your model, but not as something you can and should always fix. That's actually the right way, I think, to look at most of our validation tools — as Jane Richardson always says, something that comes up as unusual could be wrong, or it could be interesting! Best wishes, Randy Read > On 9 Feb 2016, at 20:23, Schnicker, Nicholas J <[log in to unmask]> wrote: > > Hi, > > I would appreciate some recommendations on trying to deal with RSRZ outliers as reported in PDB validation reports. I have a number of residues that visually fit the electron density fine but are still flagged as RSRZ (>2) outliers. > > I have 2 molecules in the ASU which superimpose well on one another but chain B has many more outliers than chain A. Are there things I can try to improve for a residue RSRZ score even when it seems to fit the density? > > Your input would be appreciated! > > Nick ------ Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: + 44 1223 336500 Wellcome Trust/MRC Building Fax: + 44 1223 336827 Hills Road E-mail: [log in to unmask] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk