usage: prog [options] <raw particle file> <class mx> <projections> This program can use either a 3-D binary mask, or a pair of liganded/unliganded volumes to classify particle data into 2 (or 4) groups. e2refine.py must be run on the data first. The method using a pair of volumes is much more accurate in separating particles. This is part of a multi-step process for separating ligand bound from ligand free particles, but relies on sufficient ligand/no-ligand contrast in individual images: 1) refine the entire data set with e2refine.py, such that most classes have at least 10-20 particles in them 2) construct a volume the same size as your reconstruction containing a binary mask, with 1 in the region where the ligand would be 2a) -or- prepare 2 volume files representing the particle with or without associated ligand 3) run this program <raw particle file> should be the same file used in the refinement <class mx> is one of the classification matrix files from the refinement <projections> contains the projections used for class mx Typical usage: e2classifyligand.py sets/myset_even.lst refine_01/classmx_04_even.hdf refine_01/projections_04_even.hdf --ref1 ref3d1.hdf --ref2 ref3d2.hdf --cmp=ccc --plotout=cmp.txt --pairmask --splitparticles -v 1
| Option | Type | Description |
|---|---|---|
| --version | None | show program's version number and exit |
| --verbose, -v | int | verbose level [0-9], higher number means higher level of verboseness |
| --ref1 | str | Rather than using a mask, ref1/ref2 permit using a pair of volumes for classification. |
| --ref2 | str | Rather than using a mask, ref1/ref2 permit using a pair of volumes for classification. |
| --pairmask | None | Will use the ref1/ref2 pair to generate a mask which is applied after subtracting ref1 from the particle |
| --alistacks | float | If sum of cmp results is less than the spefified value, will save the aligned particle to a per-class stack |
| --cmp | str | The name of a 'cmp' to be used when pairmask is not specified |
| --process | str | A processor to apply to the particle data before classifying |
| --plotout | str | Name of a text file for the classification plot. |
| --badgroup | None | Split the data into 4 groups rather than 2. The extra two groups contain particles more likely to be bad. |
| --badqualsig | float | When identifying 'bad' particles, particles with similarities >mean+sigma*badqualsig will be considered bad. Default 0.5 |
| --badsepsig | float | When identifying 'bad' particles, if s1/s2 are the similarities to reference 1/2, then those where |s1-s2| < sigma*badsepsig will be excluded. Default 0.25 |
| --postfix | str | This string will be appended to each set name to help differentiate the results from multiple runs |
| --splitparticles | None | Specify this to write new files containing the classified particles |
| --tstcls | int | Will generate tst.hdf containing test images for a specified class-number |
| --debug | None | Enable debugging mode with verbose output and image display. Not suitable for real runs. |
| --ppid | int | Set the PID of the parent process, used for cross platform PPID |
For more information go to emanwiki/EMAN2/Programs/e2classifyligand.