Actions
Feature #1175
closedFix Xmipp 3d refinement
Status:
Closed
Priority:
Normal
Assignee:
Category:
Image Processing
Target version:
Start date:
02/07/2011
Due date:
% Done:
0%
Estimated time:
Description
- is ist a feature or a bug ?
getting Xmipp back into the pipeline
Updated by Arne Moeller almost 14 years ago
submitted working xmipprefine.py to trunk to be able to test the webpages
Updated by Arne Moeller almost 14 years ago
- Status changed from New to In Code Review
- Assignee changed from Arne Moeller to Lauren Fisher
- Target version set to Appion/Leginon Future Version
So for me it worked on betamyamiweb. The only thing that was missing is menuprocessing - I commited that too.
One should be able to submit and upload xmippjobs from the webpage.
There are still some bugs involved (resolution, download of the map and import parameters)in this submission which I am trying to fix.
Updated by Lauren Fisher almost 14 years ago
Arne,
Most everything in the code looks good, but I have a few comments/questions:
- Xmipp Job Generator page 1 says EMAN Job Generator. This should be changed to a generic title such as Recon Job Generator.
- A lot of the help info on Xmipp Job Generator page 2 is either not helpful or is missing.
- I think the rundir should be the entire path, not '.', it seems to me this could cause a bug at some point. I have edited runXmippRefineJobGen.php in my sandbox to fix this, if you want the code.
- I know there was a bug with the threads so they have to be equal to '1', but is there a reason '1' is hardcoded in the both php (
$ejob.='--NumberOfThreads=1 '
) and python (protocolPrm["NumberOfThreads"] = self.params['alwaysone']
) scripts? I think it should only be set in the php code and then the python should setprotocolPrm["NumberOfThreads"] = self.params['numberofthreads']
. - What is the difference between the jobtypes
'xmipprefine'
and'xmipprecon'
in menuprocessing.php? - It would be a nice feature if page 2 of Xmipp Job Generator has the stack and initial model shown at the bottom the same way EMAN does.
- When uploading the final recon, there should be a php error check on the Snapshot Option 'Mass'. I thought it was optional and then got a fatal error in the log saying
*** FATAL ERROR *** mass was not defined
. If it's mandatory I don't think it should let you submit without it. - Finally, there is a bug with the icosahedral symmetry when uploading. Symmetry is set to 'i1', but in uploadXmippRecon.php it calls apSymmetry which thinks icosahedral should be 'i'. Maybe the line in apSymmetry.getSymmetryDataFromName should read
if symtext "i" or symtext "i1":
Below is the error message I got when trying to upload:... Looking up symmetry for input: i1 Traceback (most recent call last): File "/opt/myamisnap/bin/uploadXmippRefine.py", line 238, in ? upload3D.start() File "/opt/myamisnap/bin/uploadXmippRefine.py", line 98, in start fixedq['symmetryGroup'] = apSymmetry.findSymmetry(protocolPrm["SymmetryGroup"]) File "/opt/myamisnap/lib/appionlib/apSymmetry.py", line 19, in findSymmetry symdata = getSymmetryDataFromName(symtext, msg) File "/opt/myamisnap/lib/appionlib/apSymmetry.py", line 35, in getSymmetryDataFromName apDisplay.printError("No symmetry named %s was found"%(symtext)) File "/opt/myamisnap/lib/appionlib/apDisplay.py", line 56, in printError raise Exception, colorString("\n *** FATAL ERROR ***\n"+text+"\n\a","red") Exception: *** FATAL ERROR *** No symmetry named i1 was found
Other than these few things, xmipp ran fine through appion on a small virus dataset. I just wasn't able to upload because of the icosahedral symmetry.
Updated by Anchi Cheng almost 7 years ago
- Status changed from In Code Review to Closed
- Target version changed from Appion/Leginon Future Version to Appion/Leginon 3.1.0
worked fine three years ago.
Actions