Hi Eike,
the EMAN2 list would probably be a better place to ask specific questions like this. Having said that...

There shouldn't be a need to do any filtration or downsampling when using the rotate_translate_3d_tree aligner, as it already does progressive downsampling/filtration as part of its normal operation. 

You could also specify symmetry,  ie   --align rotate_translate_3d_tree:sym=icos 
if the structure is symmetric, but this should primarily impact speed rather than final alignment.

Is the alignment error small or large, and is it rotational or translational?  Honestly, to debug, at the very least would need to see screenshots showing the misalignment. Ideally, access to the 2 volumes which don't align properly would be the most useful...

cheers
On Nov 20, 2017, at 3:10 AM, Schulz, Eike-Christian <[log in to unmask]> wrote:

Dear all,

I am struggling to align my low-resolution virus structure to a reference in EMAN2.


e2proc3d.py  input.mrc output.mrc --align rotate_translate_3d_tree --alignref ref.mrc


Works of course well for EMDB structures, but if I try to align my virus the alignment seems out of register. I could imagine that some internal structure leads to this effect.

Is there a way to filter the alignment to the low-resolution only?
Would you suggest a different procedure ?


Thanks a lot in advance and best regards,

Eike







--------------------------------------------------------------------------------------
Steven Ludtke, Ph.D.
Charles C. Bell, Jr. Professor of Structural Biology
Dept. of Biochemistry and Molecular Biology                      (www.bcm.edu/biochem)
Co-Director CIBR Center                                    (www.bcm.edu/research/cibr)
Co-Director National Center For Macromolecular Imaging                  (ncmi.bcm.edu)
Baylor College of Medicine                             
[log in to unmask]