Hello Breylon Riley:
We routinely extract the particles using the box files generated from other programs like e2boxer.py or appion. Just make sure that the box files are in the same folder as the images and having the same names as well but with .box extension. Particle extraction step is pretty fast, so usually we don't use the MPI mode. Please try once without mpi and maybe you can get the correct output. Since you just pick up small number of the images, it is common that relion gives you the warning message about missing coordinate files for the images you haven't picked.
Xiaodi
________________________________________
From: Collaborative Computational Project in Electron cryo-Microscopy [[log in to unmask]] On Behalf Of Breylon Riley [[log in to unmask]]
Sent: Monday, November 09, 2015 11:16 PM
To: [log in to unmask]
Subject: [ccpem] Particle extraction step: src/image.cpp line: 80 error
Hello all,
I'm going through the relion tutorial and trying to complete the Extracting and normalising particles step.
In the previous Manual particle picking procedure, I used e2boxer.py to pick and generate coordinate (.box) files. Out of 571 micrographs, I picked from 19 to get ~3000 to use as my class 2d average template.
In my current efforts to extract and normalise particles, I'm following the tutorial closely except deviating from the Coordinate-file suffix and Particle box size parameters. Otherwise everything else is identical. But when I try to run the extraction process via the gui or script, I get
File: src/image.cpp line: 80
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 14 in communicator MPI_COMM_WORLD
with errorcode 1.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
When the run completes, I get a new Particles directory (not /Particles/Micrographs) with directories for each of the 19 coordinate files (all of which are empty). Additionally, I don't get a particles_manual.star output file despite naming it in the Extract rootname (which backs up the idea that nothing is getting extracted).
Perhaps the MPI error is a server-related issue but I wanted to ask. And as the other hiccups are concerned (e.g., missing particles_manual.star, etc) are these a consequence of using e2boxer.py and missing an intermediate step before attempting an extraction?
Thank you very, very much for you help!
(P.S., perhaps it would behoove me to note that I also get warning messages for "missing coordinate files" of the micrographs I haven't picked Though I would fully expect that, perhaps this is a sources of error that I've overlooked).
--
Best,
Breylon Riley
|