nmquijada
commented
4 years ago Hi Brad,
Really thanks for sharing your code. I think it will be helpful for other users using HPC and slurm. When we started the tormes project, networking with people by giving and receiving feedback was one of our main aims. Have you considered creating a GitHub repository? If so, I would reference to your code in the tormes documentation. Otherwise I might refer to this issue. Good that you opened the box of scripting! The first most important thing is that it gets the job done :)
Best, Narciso
biobrad
Hi all, not an issue, but thought I would share a script I wrote that auto-creates the samples_metadata file from the files you have in your folder and also adds the reference genome file to the arguments. It then sftps your files over to a HPC (it is configured for the one at my university so you would need to change it). It also creates a 'slurm' job submission file to run tormes. Finally, it also creates a script to retrieve the files once you receive notification that tormes has finished.
It is my first script so it is probably a bit under optimised, but it gets the job done and the parts that parse the file names into an array and then puts them into the samples_metadata text file may come in handy for other people (that was the part that took me the longest!!!)
Cheers! Brad tormesbot.txt