[Chimera-users] em fit refinement

Tom Goddard goddard at sonic.net
Wed Aug 15 09:40:53 PDT 2012


Hi Bala,

   Chimera fitting does rigid body rotation and translation to optimize 
the correlation of a map simulated from your molecular model to the EM 
map.  Once you've done that there is no further "refinement" that 
Chimera can do.  It does not have any flexible fitting methods.  You can 
however get different fits by using different fitting parameters in 
Chimera and it is rarely clear which set of parameters is best.  You can 
vary the region of the EM map that is used by changing the contour 
level.  Only data above the displayed contour level is used.  You can 
change the simulated resolution for you molecular model.  Another 
"refinement" to the fit is to take into account clashes with neighboring 
models.  That can be done with the "fit" command symmetric fitting 
option (if the map has symmetry) or sequential fitting option if you 
have more than one molecular model fit into an asymmetric map.

   Video examples of the different types of fitting are on the Chimera 
video tutorials web page.

     http://www.cgl.ucsf.edu/chimera/videodoc/videodoc.html

   To use other programs to do say flexible fitting you would write out 
your initial fit done in Chimera with File / Save PDB... and make sure 
to check the "save relative to" option and choose the map you were 
fitting into.  This saves the PDB in its new position in the coordinate 
system of the map.  That way other software loading that PDB will place 
it in the location you found by fitting in Chimera.

     Tom

> Friends,
> I used "fit" command to fit a structure onto a EM map and obtained a
> list of fits. Now i want to choose the best fit and refine it further.
> I would like to know if there is any way to refine a fit produced by
> fit command in Chimera. Or can i use any other software for the same,
> which can take chimera's fit as input and refine it further.
>
> Thanks,
> Bala
>
> --
> C. Balasubramanian
>




More information about the Chimera-users mailing list