Hi Dmitry,
I haven't looked back into the code, but it could be that the auto-refine
part of 2D is only implemented as the automated decrease of angular
sampling, coupled with a convergence criterion. It could be that the
--split_random_halves is not implemented for 2D. That would mean that also
postprocessing etc is currently not possible.
HTH,
Sjors
> Dear colleagues,
>
> We are very much interested in the possibility of doing an auto-refinement
> on a 2D class average, as suggested by Sjors in his reponse to a post from
> Matthew Iadanza.
> I used the (2D) class average as a reference and the particle projections
> belonging to that class as input. According to Sjors the 2D reference
> should signal the refinement program to perform a 2D refinement. However,
> I got the following error message:
>
> BackProjector::getLowResDataAndWeight%%ERROR: only implemented for 3D
> case....
>
> File: src/backprojector.cpp line: 578
>
>
> Any idea what goes wrong?
> I am using Relion version 1.4.
>
> A follow-up question: Is it possible to do a 2D post-processing, like you
> would do in 3D (and get the final resolution and FRC plot, and the option
> to do B-factor sharpening and MTF correction)?
>
> Sincerely,
>
> Dmitry
>
--
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|