[Chimera-users] Accuracy of helical parameters

Tom Goddard goddard at sonic.net
Mon Apr 20 11:37:34 PDT 2020

Hi Vadim,

  The Chimera "measure symmetry" command determines the helical rise and twist by fitting the helical density map to a rotated and shifted copy of itself when you use the optimize option.


Finding how the rotated map copy fits to the unrotated copy uses the fitmap command and should be very close to the exact best fit.

	https://www.cgl.ucsf.edu/chimera/current/docs/UsersGuide/midas/fitmap.html <https://www.cgl.ucsf.edu/chimera/current/docs/UsersGuide/midas/fitmap.html>

The docs say the fitmap convergence criteria (gridStepMin) is 0.01 times the grid spacing.  You could check how reproducible a value you get by taking two copies of the map, rotating one and fitting it several times and see what variation you observe in the numbers.  If you can it might be worth seeing whether the length of your helical map affects the result since there may be end effects where the two copies of the map don't overlap each other.

  The real accuracy of your rise and twist parameters cannot be seen by looking at the reconstructed map.  You need to ask the software that made the helical reconstruction from the single particle images.  It optimized the rise and twist parameters when it constructed the 3D map from 2D images and that is likely to be the biggest source of uncertainty and errors in the parameters.  To gauge that you might reconstruct the filament multiple times, maybe scrambling the ordering of the input maps to get different reconstructions and see how much the parameters vary.


> On Apr 20, 2020, at 2:31 AM, Vadim Kotov <vadim.r.kotov at gmail.com> wrote:
> Dear mailing list,
>  I have a question regarding command "measure symmetry" in helix mode. Is there a way to estimate the accuracy for the obtained helical twist and rise? My guess is that the accuracy is related to the pixel size, so if I have a map with a pixel size of 1 A, then the standard deviation for helical rise could be 0.5 A.
> Many thanks!
> Vadim Kotov
> Postdoctoral fellow, Löw lab
> EMBL Hamburg
> _______________________________________________
> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu
> Manage subscription: 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/20200420/a124169c/attachment.html>

More information about the Chimera-users mailing list