-
Notifications
You must be signed in to change notification settings - Fork 4
Frequently asked questions
- What is the format of the CMM file used to provide a vector defining a subunit of interest?
- What does --align_subparticles option do?
- Is there an easy way to define 5-fold (or 2-fold or 3-fold) axis of an icosahedrally symmetric particle?
- How does localized reconstruction differ from other methods such as focussed classification?
- Why do I get an pyworkflow.utils error?
- The subtraction seems not to work, I can see a shadow of my particle
The CMM file (Chimera marker file) should have the following format. The first marker defines the start point of the vector (typically in the origin of the particle reconstruction) and the second marker defines the position of the subunit.
<marker_set name="marker set 1">
<marker id="1" x="0.000" y="0.000" z="0.000" r="1" g="1" b="0" radius="10"/>
<marker id="2" x="110.000" y="0.000" z="40.000" r="1" g="1" b="0" radius="10"/>
</marker_set>
It is a good idea to open the CMM file and your particle reconstruction in Chimera to see that they match as intended.
Option --align_subparticles will align the vector defining the orientation of the subunit on the Z-axis. For example, if the vector defines a 5-fold vertex of icosahedrally symmetric virus, this option will align the 5-fold axis of the reconstructed subparticle map on Z.
Is there an easy way to define 5-fold (or 2-fold or 3-fold) axis of an icosahedrally symmetric particle?
To create subparticles for subunits at icosahedral symmetry axis, use these vectors (this assumes the so called "I1 orientation"):
- all 2-folds, use --vector 0.000,0.000,1.000
- all 3-folds, use --vector 0.382,0.000,1.000
- all 5-folds, use --vector 0.000,0.618,1.000
How does localized reconstruction differ from other related methods such as focussed classification?
The key difference in the localized reconstruction method is that several smaller areas (subparticles) are extracted from the original particle images. This allows combining several substructures from a larger complex (60 in case of an icosahedrally symmetric assembly such as a virus, but lower symmetries are possible). Subparticles inherit their starting orientations from the particle they originated from, but the orientations can be further refined, around the new origin. Focussed classification methods focus on only one local area within the particle structure, defined with a mask that is applied during 3D classification. In this method the rotations are relative to the particle origin and typically alignment is thus skipped in classification. Both approaches are typically combined with partial signal subtraction (Bai et al. 2015 eLife) and can benefit from relaxing particle symmetry either using the script provided relion_create_symmetry_related_particles.py (or relion_particle_symmetry_expand as part of Relion 2.0) (Scheres 2016 Methods Enzymol).
If you get this error
import pyworkflow.utils as pwutils
ImportError: No module named pyworkflow.utils
It probably means you forgot to add scipion run in front of the script command.
If you are using a map after postprocessing (with inverse B-factor sharpening), try using the map that was produced by Refine3D or relion_reconstruct instead. The values (amplitudes) in the map after post processing are scaled differently to the values in your data.