[Chimera-users] resolution for volume fitting

[Tom Goddard]

Hi Bala,

   One idea is to set the density map contour level so that the surface 
encloses the expected volume for the molecular assembly.  An example of 
that is in a Chimera tutorial

     
http://www.cgl.ucsf.edu/chimera/data/tutorials/eman07/chimera-eman-2007.html#part2

Just the data inside the contour level is used for computing correlation 
coefficients in Chimera (unless "use only data above contour level..." 
option is turned off).  So the contour level has a big effect on the fit 
correlation coefficients, for example it might vary from 0.5 for a low 
contour level to 0.9 for a high contour level.  It doesn't mean the fits 
are any different -- they could be in identical positions.  But the data 
used for calculating the correlation is different.  People rarely talk 
about this but basically you can get any correlation value you want.  
Which contour level is best.  There is no obvious answer to that.  But 
the suggestion above to use the level to enclose the right volume is 
sensible to try.

     Tom


> Hi Elaine,
>
> Thank you.
>
> I wanted to ask about choosing an optimal threshold level of the 
> volume data (the experimental map) instead of which i wrote as 
> resolution of the volume data. I admit that i have not mentioned it 
> clearly in the previous mail.
>
> I tried to fit a structure on to a map and i found that the 
> correlation value, Avg and Inside in the fit-list changes depending up 
> on the threshold level of the volume data i chose. So i am not getting 
> how to choose a correct threshold value of the volume data.
>
> Cheers,
> Bala
>
>
> On Wed, Mar 7, 2012 at 7:01 PM, Elaine Meng <meng at cgl.ucsf.edu 
> <mailto:meng at cgl.ucsf.edu>> wrote:
>
>     Hi Bala,
>     Since you can't control the resolution of the experimental density
>     map, I'm guessing you mean what resolution value to use for the
>     map simulated from the atoms that are being fit into the map
>     (simulating a second map allows scoring the fit by map-map
>     correlation).
>
>     I believe you would want to use the same or similar value to the
>     experimental map's resolution.  In turn, that value should be
>     available from the database where you got the map and/or the
>     publication describing the map (or if not publicly available, the
>     scientists who did the experiment).
>
>     I hope this helps,
>     Elaine
>     -----
>     Elaine C. Meng, Ph.D.
>     UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
>     Department of Pharmaceutical Chemistry
>     University of California, San Francisco
>
>
>     On Mar 7, 2012, at 3:22 AM, Bala subramanian wrote:
>
>     > Friends,
>     >
>     > I am a newbie to fitting molecular structures on a volume map. I
>     would appreciate if someone could suggest me some guidelines to
>     choose an optimal resolution of the map before fitting.
>     >
>     > Thanks,
>     > Bala
>     >
>     > --
>     > C. Balasubramanian
>
>
>
>
> -- 
> C. Balasubramanian
>
>
>
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20120307/e7908597/attachment.html>