My lab has happily used the CloudMap pipeline to identify the location of mutations from in C. elegans strains using the “CloudMap Hawaiian Variant Mapping with WGS and Variant Calling” workflow. However, our most recent batch of analyses has run into an issue. When we run the data the workflow stops at the "Realigner Target Creator (GATK intervals)”, with a tool error: “failure preparing job”.
I initially thought that maybe there was something wrong with our imported workflow, so I reimported the CloudMap workflow, and ran the analysis again, getting the same error. Same results with a dataset from one of our other mutants.
Finally, I imported the ot266 'proof-of-principle" dataset and ran the analysis again. Same error. I have now tried this analysis with all available CloudMap workflows on the Galaxy main server and always get the same error at Realigner Target Creator (GATK intervals). I can only conclude that there is some issue with the GATK tool on the main public instance of Galaxy.
Can anyone suggest a remedy? I am using the datasets associated with CloudMap on the main Galaxy server and these have work up until this latest batch. I’m frankly surprised that no one else has picked up this issue. Maybe I’m missing something simple, but it’s very frustrating.
I know it’s not immediately helpful, but how about contacting Gregory Minevich directly? I think you might get some feedback quicker than on the forum to such a specific question.
Yes, I did that too. He’s looking into it. Apparently someone else reported the same problem to him yesterday. I was wondering if anyone else had experienced the same problem, or whether most people were using their own Galaxy instance.
As a bit of advertisement:
as its main author I’d like to mention MiModD here, which can do all the upstream steps of mapping and variant calling similar to what’s included in the CloudMap workflow for you on your local computer
(well, if you have a Mac or a Linux machine with 8 GB of RAM). It can produce output ready to feed into the CloudMap Hawaiian Mapping tool so you can bypass the whole rest of the workflow.
We had a report on it in the Wormnbreeder’s gazette last spring
(http://54.165.141.74/wbg/articles/volume-20-number-1/mimodd-–-mutation-identification-from-whole-genome-sequencing-data-on-desktop-pcs/)
and have improved the software quite a bit since then. The latest release (from today !) can be found at https://sourceforge.net/projects/mimodd and there is a manual (still needs a bit of updating) at http://mimodd.readthedocs.org .
We have bundles of MiModD with a local Galaxy available on the download site, which you can just unpack and run if you are using OS X or Ubuntu.
If you try it, let us know if it works for you and what you would like to have improved.
Have a nice weekend,
Wolfgang (from the Baumeister lab in Freiburg, Germany)
Just a note for anyone that stumbles across this thread, I highly recommend MiModD as a method for anyone wanting to use WGS data to map mutations in C. elegans. It’s very straightforward and we now routinely use this approach. There is an excellent tutorial, and the author was very helpful in sorting out the problems that we encountered the first time we used it. http://mimodd.readthedocs.org/en/latest/tutorial.html