warning for preoriented subtomogram extraction

[Alister Burt]

Hi all,

I just spent a while helping a friend debug some problems and thought a general message to raise awareness would be useful.

If you plan to refine particle positions and orientations after extraction in Warp, do not use the option for preorienting your particles when exporting subtomograms. 

If you do, the connection to the tomogram reference frame is lost and you will not be able to reextract your particles with the correct orientations/perform any refinements properly in M. This is because the euler angles in the star file are now parameters for the transformation of a prealigned particle to align it with the reference, rather than the transformation of the particle in the tomogram to align it with the reference.

Cheers,

Alister

[teg...@gmail.com]

That's a good point! Ironically, I added this option for easier debugging (can average sub-tomos in Matlab without going through relion_reconstruct), and left it there in case someone else would like to test something without bothering with interpolation. Maybe it's time to remove it or make it less visible.

[Alister Burt]

I think having the option is definitely useful! Less visibility or a warning in the user guide as for sparse ctf volumes seems appropriate 

Cheers,

--
You received this message because you are subscribed to the Google Groups "Warp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to warp-em+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/warp-em/42dba585-c062-4dbd-ba1d-5f2702f447d9n%40googlegroups.com.