NCS B-factor differences

[GERARD@oldvax.bmc.uu.se (Reply to: GERARD@XRAY.BMC.UU.SE and *not* to the @oldvax address)]

***  For details on how to be removed from this list visit the  ***
***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***


busy with the embo bioinformatics course so this is a tad tardy

>You know, after all these years in xtallography I'm tempted just to point
>out that for anything anyone says, there must be at least 1 counter
>example, and often more like 10 of them.

errrr, i know of at least one case where this is not true ...

but more to the point:

- to 'prove' that the average Bs are real, do the
  following: set all B-factors equal to the average
  over the complete structure; then refine grouped
  Bs in xplor or cns where every group is one of
  your six molecules. what you should see is that
  the delta B you get is negative for 4 of them and
  positive for the other two; in adition, you should
  see a clear drop in r and rfree. see:
  http://alpha2.bmc.uu.se/gerard/gmrp/gmrp.html
  (search for "P2" in that page; the first
  occurrence is in the paragraph that discusses
  the exact same problem we had with P2 myelin protein

- if you have an experimental map, you can also find
  out if or show that your molecules have different
  average Bs by comparing the densities for the
  various copies with the program SPANCSI (part
  of RAVE; manual lives at:
  http://alpha2.bmc.uu.se/usf/spancsi_man.html )

hope this helps

--cd