search

EBI-Metagenomics / MGnifyR

R package for searching, downloading and analysis of EBI MGnify metagenomics data
https://ebi-metagenomics.github.io/MGnifyR/
Artistic License 2.0
19 stars 10 forks source link

MGnify course material #47

Closed TuomasBorman closed 2 months ago

TuomasBorman commented 2 months ago

Hello,

the MGnify course is approaching.

  1. We cannot use notebook servers

The EBI's notebook servers cannot handle >30 participants simultaneously. That is why we cannot use it in a course. Also that means that there is no need to get the material to there before the course. We can add this notebook in their jupyter notebook server after the course.

  1. In the course, we use course VMs

The course organizers provide virtual machines. They have all requires packages installed. Moreover, they include memory cache so the data importing takes just around 1 minute (instead of an hour).

If you want to run the code, I can provide you the cache directory.

3, The plan of our session

We follow this vignette in the course by copy-pasting code to VM. In the first 45 minutes, I will introduce MGnifyR package and basics of TreeSE if I have time. You can use around 1 hour and 15 minutes to:

Then there should be still around 30 minutes for participants to explore OMA and have discussion.

  1. See the workflow

The workflow aims to have only necessary stuff and link to OMA. In an attachment, there is rendered workflow file.

@antagomir

SandyRogers commented 2 months ago

Just FYI I did, yesterday, get a build of the notebook server working with the new MGnifyR (i.e. R=4.4, Bioconductor 3.19 etc) whilst avoiding anaconda default channels. But it is a big change in the build process so I need to test it more, and also rewrite the existing MGnifyR-based notebooks to avoid all the deprecation warnings introduced by the changes to MGnifyR. So that will be in the coming weeks.

TuomasBorman commented 2 months ago

Just FYI I did, yesterday, get a build of the notebook server working with the new MGnifyR (i.e. R=4.4, Bioconductor 3.19 etc) whilst avoiding anaconda default channels. But it is a big change in the build process so I need to test it more, and also rewrite the existing MGnifyR-based notebooks to avoid all the deprecation warnings introduced by the changes to MGnifyR. So that will be in the coming weeks.

Thanks for the information. Bioconductor has the next release in late October. We had lots of changes in the summer in miaverse to finalize it. Now the framework is stabilized. The stabilized version becomes release version in October.

antagomir commented 2 months ago

Seems feasible to me. I will see if I have anything to add but I think this could be merged.