add UMAP script

[rjcorb]

Purpose/implementation Section

What scientific question is your analysis addressing?

Closes #569. This PR adds a script to molecular-subtyping-MB/ to generate a UMAP of MB SHH samples based on most variable methylation array probes.

What was your approach?

adapted from this code: https://jharenza.github.io/mb-subtypes/explore-medulloblastoma-unsupervised-clustering.nb.html

What GitHub issue does your pull request address?

#569

Directions for reviewers. Tell potential reviewers what kind of feedback you are soliciting.

Which areas should receive a particularly close look?

Check code and ensure script runs successfully:

R -e "rmarkdown::render('06-mb-shh-umap.Rmd')"

Is there anything that you want to discuss further?

I have the UMAP points sized by age range currently, to add some additional info on outliers and un-subtyped MB SHH samples. We may not want to keep this in the final figure.

Is the analysis in a mature enough form that the resulting figure(s) and/or table(s) are ready for review?

Yes

Results

What types of results are included (e.g., table, figure)?

UMAP

What is your summary of the results?

Overall we see good clustering of MB SHH samples into the four subtypes. There are a few samples not cluttering with their subtype, including a few adolescent/adult samples with the infantile MB SHH samples. We may need to review these samples to check how they were classified into subtypes.

Reproducibility Checklist

• The dependencies required to run the code in this pull request have been added to the project Dockerfile.
• This analysis has been added to continuous integration.

Documentation Checklist

• This analysis module has a README and it is up to date.
• This analysis is recorded in the table in analyses/README.md and the entry is up to date.
• The analytical code is documented and contains comments.

[jharenza]

hi @rjcorb can you stack the MB PRs on #571? I had to remove MM2S code there.

[zzgeng]

This night be a typo, do you mean 20k?

[jharenza]

@rjcorb can you do me a favor and label the following samples on the UMAP? These have very high-conf non-MB histologies and I wonder if these are some of the problematic samples.
7316-1884
7316-1676
7316-1666
7316-3202

[jharenza]

Can you also create an mb umap by the 4 subtypes with the expression data?