|
⇤ ← Revision 1 as of 2007-04-11 14:27:05
Size: 271
Comment:
|
Size: 1080
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| Q: I need to do a 2D average with symmetry using all my particles. I could not | '''Question:''' I need to do a 2D average with symmetry using all my particles. I could not |
| Line 5: | Line 6: |
'''Answer:''' Are you saying that you have a set of particles all in the same orientation, and you wish to align them, then impose symmetry ? If so, there are several approaches. Assume your particles are in ptcl.hed/img, and you have a 4-fold symmetry: 1. The easy approach: {{{ classalign2 ptcl.hed 5 keep=5 refine proc2d classes.hed sym.hed sym=c4 }}} If you wish to discard some of the worst particles, lower keep=5 to, say keep=1. 1.#2 A better approach: {{{ refine2d.py ptcl.hed --iter=8 --ninitcls=6 proc2d iter.final.hed sym.hed center proc2d sym.hed sym.hed inplace sym=c4 }}} This will subclassify your particles into 6 groups, and make an average for each group. You can then compare iter.final.hed to sym.hed and see if the imposed symmetry is justified. |
Question: I need to do a 2D average with symmetry using all my particles. I could not find a command in the manual. "startnrclasses" is no symmetry. And "startcsym" is for 3D reconstruction and could only use 1/3 of my particles. Is there such a command in EMAN? Thank you.
Answer: Are you saying that you have a set of particles all in the same orientation, and you wish to align them, then impose symmetry ?
If so, there are several approaches. Assume your particles are in ptcl.hed/img, and you have a 4-fold symmetry:
- The easy approach:
{{{ classalign2 ptcl.hed 5 keep=5 refine proc2d classes.hed sym.hed sym=c4 }}}
If you wish to discard some of the worst particles, lower keep=5 to, say keep=1.
- A better approach:
{{{ refine2d.py ptcl.hed --iter=8 --ninitcls=6 proc2d iter.final.hed sym.hed center proc2d sym.hed sym.hed inplace sym=c4 }}}
This will subclassify your particles into 6 groups, and make an average for each group. You can then compare iter.final.hed to sym.hed and see if the imposed symmetry is justified.