Please not that average B value may be arbitrary depending on data processing program, reference image etc. Average B values could be arbitrary (you can add to all atom single B value and it will not change information content of the data, they are removable). I think either variance of B or <B> - Bmin may correlate with resolution better.
regards Garib On 16 May 2012, at 15:06, Nat Echols wrote: > On Wed, May 16, 2012 at 6:46 AM, Qiang Chen <c...@red.dfci.harvard.edu> wrote: >> I have a 2.4A structure(pdb code 3LAF)with an average protein b-factor of >> 48. I wonder whether it's acceptable. Is there a direct correlation of >> b-factor and resolution? > > They're correlated, but it's not an exact relationship. See attached > plot (which includes all entries in the PDB for which structure > factors were deposited, as of about a year ago). The mean B-factor > for structures between 2.3A and 2.5A resolution is just under 40, so > your structure is very reasonable (the distribution is quite wide). > >> The R and Rfree are 21.1% and 23.1%, respectively. >> This structure has a very high solvent content, 75%. Does it affect the >> b-factors? > > Only to the extent that high solvent content tends to result in more > poorly ordered crystals and worse resolution - also, perhaps surface > side chains will be more flexible if they're not forming crystal > contacts. But it doesn't inherently change anything in refinement, > aside from making it easier because you have a better > observations-to-parameters ration. > > -Nat > <B_versus_d_min.png> Dr Garib N Murshudov Group Leader, MRC Laboratory of Molecular Biology Hills Road Cambridge CB2 0QH UK Email: ga...@mrc-lmb.cam.ac.uk Web http://www.mrc-lmb.cam.ac.uk
