Hi Cynthia,
being a novice in Relion, I am also trying to reproduce the results
described in "Amyloid structure determination in RELION-3.1" by Sjors.
The best I can get for the "second set" of particles (256 pixels without
down-scaling) is 12 classes, still way less than what stated in the
paper. Attached please find a detailed description of my procedure.
Probably not that helpful to you but at least it can serve as a
reference for more helpful discussion. What I didn't describe in the
attached file is that one can get a lot more 2d classes if one chooses
"Yes" for "Ignore CTFs until first peak?" in the CTF tab. However, the
results are similar to your second screen shot - the images do not have
any resolved beta-sheet spacing. I did try to create an initial model
but the resultant 3D map has a lot of empty space. In principle we
should try different inter-box distance, regularization parameter,
in-plane angular sampling rate, tube diameter, and etc. But I prefer to
use the parameters suggested in the paper because the key issue is
whether we can reproduce the results reported there. Hope that some
experts out there, or Sjors, could give us some hints.
Best,
Jerry
--
Jerry C. C. Chan
Department of Chemistry
National Taiwan University
Taiwan
http://spin.ch.ntu.edu.tw
886-2-33662994
########################################################################
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/
|