[3dem] Fixing pre-alignments
Bill & Sue Tivol
wtivol at sbcglobal.net
Wed Nov 5 17:39:24 PST 2014
On Nov 5, 2014, at 4:49 AM, Mike Strauss wrote:
> once you fix the alignment in Midas, and save the transformations,
> you just need to rerun “Generate Course Aligned Stack” ( think
> that’s what it’s called). Do NOT run the automatic correlation step
> again, as that will overwrite your Midas work.
> As I understand it, the process is as follows:
> -find frame-to-frame transformations (either by correlation or by
> Midas) - these are saved as a .xf file
> -calculate the global transformations (to retain the majority of the
> image in the frame) - using xftoxg
> -apply global transformations to the unaligned stack
> Etomo takes care of the second two steps at once with the “Generate
> Course Aligned Stack” button.
> If you want to try to get better initial automated alignment, it may
> be worth trying any (combination) of the following:
> in the advanced tab:
> Click Exclude central peak
> Change frequencies used
> Try No Cosine Stretch
> Try Cumulative correlation
> I’ve had some luck with some of these in the past, but it does
> depend on your data set.
Hi Ariane, Huy, Lu, & Mike,
Thanks for your responses. I think my error was rerunning the cross-
correlations. I put in Mike's response, since it was the most
complete. Thanks also for the suggestion of the IMOD list; I didn't
know there was one.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the 3dem