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


Hi Nick,

you may like to give our automated homology modelling and molecular 
replacement server a look - just to check out what it yields:

http://www.igs.cnrs-mrs.fr/Caspr/

It will run AMoRe for molrep, but it uses in addition to your templates a 
series of perturbed homology models from MODELLER as input. 

Kind regards,

Karsten.


On Tuesday 14 February 2006 05:40, Noinaj wrote:
> Hi,
>
>
> I am working on a structure of an enzyme that I recently collected a 2.4
> angstrom native dataset.  I have a search model that is 24% identity and
> 40% similar, equally distributed throughout the entire molecule. 
> Fortunately, I have been able to find very nice rotation and translation
> solutions via MOLREP (CNS), typically with top solutions being 3 or 4 times
> over background.  The spacegroup is C2 and I have 1 molecule in the
> asymmetric unit.  After rigid body refinement, I have been able to get the
> Rwork and Rfree down to ~0.44 for each, but can't seem to refine any
> further.  If I do minimize or simulated annealing, my Rwork goes down
> nicely, but the Rfree goes up, almost in equally opposite magnitude.  I
> have made maps using the current model, but the maps are still rough and
> not as nice as I had hoped.
>
> With getting nice solutions for the rotation and translation functions, I
> had hoped for better luck with refinement.  Currently I am trying several
> things to try to improve my model and running through several programs
> also.  So far, not much luck.  I am working towards obtaining experimental
> phases, but would really like to get the molrep to work out if possible.
>
> Any advice or direction for a young crystallographer?  Thanks in advance.
>
>
>
> Cheers,
> Nick

Reply via email to