very low B values

Jiang Jiansheng jiang at LAPLACE.CSB.YALE.EDU
Thu Aug 31 23:33:15 EST 1995


>      Hello.  I am working with drug/DNA complexes.  A structure that I have
>  been working with has refined to an R of .17 and a free R of .25 with data
>  from 8 to 2.25 A.  However, with NUCLSQ (a nucleic acid variant of PROLSQ),
>  I have very low B values for many non-solvent atoms, with many as low as zero.
>  Refinement
>  in XPLOR also gave me these low B values though XPLOR doesn't seem to allow
>  anything below 2.0.  I am at a loss to explain what is going on here.  Does
>  anyone have an idea?

>  --

>  Ho Leung Ng
>  University of California, Los Angeles
>  Molecular Biology Institute / School of Medicine
>  ng at tesla.mbi.ucla.edu

Inspect the X-PLOR output file, the overall B factor might had been
refined to be negative.

Check the Wilson plot - your data were probably over-corrected with
an overall temperature factor, i.e., the determined B factor from the wilson plot
might be zero or negative.   Try the following (in X-PLOR 3.1),

    evaluate ($BW=16.0)
    do  amplitude (fobs=fobs*exp(-$BW/4*s()^2)

before the B refinement, then see whether the individual B's are reasonable.

Jiansheng Jiang.




More information about the X-plor mailing list