*** For details on how to be removed from this list visit the *** *** CCP4 home page http://www.ccp4.ac.uk ***
Hi Understood and agree. Perhaps my wordings in the original post can be a bit better. Perhaps I should ask what have others *found* to be inside the hydrophobic core/region :) Anyway, in one of the structures I'm working on, there'r 2 (identical) hydrophobic cores, generated by 2mol/ASU (not NCS-averaged, yet). Each core has 2 blobs of stray density, all appear to have the same shape. They can't be H2O, neither shape nor size fit, but yet seems abit too small for MES (in the condition). Just curious what they are, that's all. It's a 1.6A data set. Anastassis Perrakis: "...'better modelling' not 'better R'.." Ok yes, better modelling :) Thanks Andrew -------------------- PhD Student Dept of Biochemistry Queen's University Kingston ON Canada On Thu, 8 Dec 2005, Dirk Kostrewa wrote: > *** For details on how to be removed from this list visit the *** > *** CCP4 home page http://www.ccp4.ac.uk *** > > > Dear Lynn Ten Eyck, > > I fully agree with you and post this reply only because of the students > that are new to crystallography. Even interpretation of the electron > density of the chemically well known protein polypeptide chain has a > wide grey area: surface side chains, flexible loops and termini, > alternate conformations, etc. (you all know that, of course!). > Interpretation of waters, ions, or any other compounds from the > crystallization buffer is even more fuzzy. But I agree with you, that > one should try the best to get both a physically and chemically > reasonable model. I must admit, however, that even with a well-refined > protein with, say, 2 A with R/Free-R ~ 15/19% and good stereochemistry, > I still have the feeling that there are errors that I didn't > detect/correct, and this is certainly true. As in the saying " . . . > protein refinement is never finished but aborted . . ." (was it you, > Phil, who said that?) > > Best regards, > > Dirk. > > Lynn Ten Eyck wrote: > > >*** For details on how to be removed from this list visit the *** > >*** CCP4 home page http://www.ccp4.ac.uk *** > > > > > >Please do *NOT* just add what are essentially content-free dummy atoms > >just to lower the R. This just adds extra parameters to soak up > >potentially valuable information about the quality of the structure. > > > >The real refinement problem is to find a physically plausible model > >that, properly adjusted, fits the data to within the experimental error > >of the data. Uninterpreted blobs are not a molecular model. If you DO > >put a molecular component on that density, you are in effect saying "I > >really believe, just as strongly as I believe the conformation of the > >main chain, that this density is (say) an acetate ion." Do you? If so, > >put it in, but you have actual chemical evidence for the composition of > >the protein chain. > > > >Best regards, > >Lynn Ten Eyck > > > > > > > > > > > -- > > **************************************** > Dirk Kostrewa > Paul Scherrer Institut > Life Sciences, OFLC/110 > CH-5232 Villigen PSI, Switzerland > Phone: +41-56-310-4722 > Fax: +41-56-310-5288 > E-mail: [EMAIL PROTECTED] > http://sb.web.psi.ch > **************************************** > >
