suggestions on refinement

Yu Wai Chen ywc at mrc-lmb.cam.ac.uk
Wed Jan 12 09:49:16 EST 2000


Balaji Bhyravbhatla wrote:
> 
> Dear All,
> I have a molecular replacement solution with initial  R's between 38 and
> 40%.
> I have been trying to refine this model against our dataset but am not
> having much
> luck.  This is what happens when I do either minimize or annealing:
> 
> REMARK coordinates from minimization refinement
> REMARK refinement resolution: 100.0 - 3.0 A
> REMARK starting r= 0.3815 free_r= 0.4102    <<<<<<<<<<<< Look <<<<<<<<
> REMARK final    r= 0.4971 free_r= 0.6025        <<<<<<<<<<<<Look
> <<<<<<<<
> REMARK rmsd bonds= 0.020111  rmsd angles=  4.89617
> REMARK wa= 9.39831
> REMARK target= mlf cycles= 10 steps= 50
> REMARK sg= P2(1) a= 26.37 b= 26.91 c= 101.92 alpha= 90 beta= 89.98
> gamma= 90
> 
> This is after a round of just minimization. When I do annealing it is
> even worse!!! and the
> r-factor jumps up to the high 50's and free_r is in the low 60's!! I

These are signs of very serious yet simple global errors.  I would
double check cell dimensions, space group etc very carefully.  Are you
sure your space group is correct?

> Can anyone suggest what I need to monitor to find the cause? Or may be
> another way of
> pushing the refinement??  Or just plain suggestions are welcome so that
> I know I am not
> missing something!!

I don't think it is wise to push refinement.  Something is seriously
wrong, that cannot be compensated by playing with refinement parameters.

> I have tried the ccp4 set also a bit (arp, refmac etc.) but no
> improvements there either.

You may want to try to use AMoRe to see if you can get the same
solution.

-- 
===================================================================
Yu Wai CHEN, Ph.D. ..................   email:ywc at mrc-lmb.cam.ac.uk
 Centre for Protein Engineering,             tel:+44-(0)1223-402148
 MRC Centre, Hills Rd, Cambridge CB2 2QH, UK fax:+44-(0)1223-402140
 WWW homepage: http://www.mrc-cpe.cam.ac.uk/~ywc




More information about the X-plor mailing list