***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***


Dear Yi,

it is very bad practice to refine an MR solution without looking at the
map after MR first, especially at low resolution. The aim of refinement is
to reduce R/ Rfree and it can do so by overfitting your data. This means
that differences between the MR solution and your data become smoothed
away, and you do want these differences in order to improve the model.

So carry out the MR, then build and correct the model as much as possible
before doing any refinement.

Tim

--
Tim Gruene
Institut fuer anorganische Chemie
Tammannstr. 4
D-37077 Goettingen

GPG Key ID = A46BEE1A


On Thu, 3 Aug 2006, Yi Xue wrote:

> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
>
>
> Dear all:
>
>
>     I have one dataset, diffracted to 3.2A.
>     It should be a protein-DNA complex. The protein structure has already
> been solved. The DNA is a 23mer.
>
>     I used phaser  to do the molecular replacement, and it found a solution,
> RFZ=7.2, TFZ=11.9, however, LLG is negative.
>
>     I used the solution, did 20 cycles of restrained refinement, using
> refmac.
>
>    R decreased from 0.476 to 0.354
>    Rfree decreased from 0.453 to 0.45.
>
>    Since there is only trivial improvement for Rfree, so is it possibly a
> right solution?
>
> thanks
> Yi
>
>

Reply via email to