Dear Rich,
You would need to refine in C1, with the quarter-rise and
close-to-90degree twist (but depending on the twist in the C4
refinement). Hopefully, if you start from a relatuvely low-pass filtered
version of the C4 model, that would still converge well. But it all
depends on your structure, so hard to tell from email. For some
filaments, refining in C1 (without the additional 4-fold averaging from
the 4-start helix; so using the same rise and twist from the C4
refinement) also works and then gives a better insight into the real
symmetry. In any case, don't rely on too large search ranges for tilt
and rise in the refinement, as that energy landscape is complicated.
Good luck,
Sjors
On 28/06/2022 12:47, Richard Collins wrote:
> Hi Sjors
>
> Yeah - I’d worked that out from the paper. I can see the C4 implementation is nearly but not quite right; If I allow a search of twist and rise, it will proceed going up and down in value and gradually drift way from the correct values producing smooth tubes that are low resolution correct but featureless.
>
> It was more the implementation of the revised and newly defined p 1_4 values I’ve been thinking.
>
> Do I take the p twist and rise values, impose these in the helical reconstruction of the existing C4/helical class and then rerun at the same time but using C1 symmetry? Or is there a specific p** symmetry operator I need to use?
>
> Cheers
>
> Rich
>
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1
This message was issued to members of www.jiscmail.ac.uk/CCPEM, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|