Hello,
I submitted a job to unicycler over 24 hours ago to the main USA galaxy server and it is still grayed out waiting to run. How long does this typically take?
I am prepping for a course I am teaching and I was going to have my students do the “Unicycler assembly of SARS-CoV-2 genome with preprocessing to remove human genome reads” tutorial and I want to ensure they have enough time to perform it.
After 4 days and no responses I purged my history and am attempting again. I also downloaded the fastq results up to that point to run locally in ubuntu. I had already attempted this before but got a spades error.
If anyone out there can help me I would greatly appreciate it. I would love my students to perform a de novo assembly.
@mharpole have seen your post here only now. Read your corresponding message in the Galaxy Training Network chat, but your issue wasn’t so clearly described there.
I cannot comment on Unicycler jobs on usegalaxy.org. There may indeed be a long queue you are in there, though 4 days is really long.
What I do know is that people have followed that very same tutorial successfully on usegalaxy.eu last week so maybe you have more luck there.
In general, however, it is a good idea when planning to do a tutorial like this with a challenging tool like Unicycler (I think the training material mentions this bottleneck explicitly) to create an example history before the actual course.
During the course let participants perform all steps up to that challenging one, but then explain them what computational complexity is hiding behind that next step and that this cannot be scaled reliably to a large number of participants. At this point, you can share your prepared history with them, they can copy the dataset they need from yours into their history, and continue.
Please also note that usegalaxy.eu offers TIaaS, which makes trainings a lot more reliable though in the end it’s always good to have a strategy for when things don’t work as they should.
2 Likes
Thank you for the response. My students are not slated to perform the analysis until next month and I was prepping an example history. I successfully performed the unicycler analysis locally, it was unhappy about the file extension fastqsanger.gz and I just removed the sanger portion. I didn’t even think about it when I downloaded the data. I appreciate the suggestion about the .eu server and I will look into TlaaS.