Hi Jun,
Since the extracted subvolumes are already aligned, you may want to try 3D classification first without prior refinement.
You can use the subvolume average from PEET as an initial reference for this, but should probably set "Ref map is on absoute grayscale" to "No". You probably also should re-extract larger aligned subvolumes before trying further 3D refinement in RELION. (By default, aligned subvolumes will exactly match the size you specified for your final average. Most likely you'll want to increase this by 1.5 - 2X).
You can also do 3D classification directly in PEET using programs pca and clusterPca. Pca would only need to be run once, and then you can try varying the number of classes and the number of features to be used in cluster Pca.
I'll be interested in hearing how your refinement and classification results in RELION compare to those from PEET. Of course, the 2 packages are optimizing different objectives, so you shouldn't expect identical results.
Regards,
-jh-
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
|