Hi all,
Has anyone else encountered the following half2-map issue in Relion 3.1 helical reconstruction:
We are doing helical reconstruction on several different filamentous samples. Particles were picked with CrYOLO and extracted in Relion 3.0 and from then on, everything was carried out in Relion 3.1. When we set 3D auto-refine with local search ranges, the algorithm works fine for both half-maps and optimizes the rise and twist inside the suggested range. However, in the final iteration, half1-map stays at the refined parameters of rise and twist, while the half2-map always jumps back to the set initial parameters, even though its values in the second-to-last iteration were different from the initial ones. After convergence, this affects the final averaged values for rise and twist and the final refinement result. Relion doesn't report any warnings or errors and the final maps converge to about ~4A and side chains can be distinguished. Is this some sort of a bug in writing out the final parameters or does this prevent the refinement from converging to the best possible result?
Thank you in advance,
Andreja Kežar, PhD
Department of Molecular Biology and Nanobiotechnology
National Institute of Chemistry
Ljubljana
Slovenia
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
|