Re: [scipion-users] Aborted Auto-refine

[Juha H. <juh...@he...>]

Hi David,

I propose to create a new protocol for refinement, then select the option
to continue from a previous refinement (you should see the interrupted
refinement as an input).  Then in the additional parameters field (the last
tab) add option --force_converge.  Then the continuation refinement will
enter the last final iteration straight away.

If you would rather import the particles using a STAR file and calculate
the half maps separately, you should check that the particle orientations
and origins, in addition to the half data set assignment, have been
imported properly. I am not sure if this happens currently, but you can
check by examining the particle set (display results) and changing to the
list view.

Best wishes,
Juha

On Mon, May 25, 2020 at 10:43 AM DAVID GIL CANTERO <dg...@cn...> wrote:

>
> Hello, I had an auto-refine running whose resolution have changed just
> 0.01A from the previous iteration to the last one (not the final). As
> the difference is not great and the iterations takes more than 2 days
> I stopped it but now, none of the protocols admit it as an input, it
> did not even appear in the inputs list. I have tried to import the
> .star file with the angles from the last iteration but when
> reconstructed, the resolution is far worst than what said in the
> refinement. I guess the problem is having aborted the refinement, is
> it any way to by-pass this issue?
>
> Thank you.
> David.
>
>
>
>
> _______________________________________________
> scipion-users mailing list
> sci...@li...
> https://lists.sourceforge.net/lists/listinfo/scipion-users
>