Add pathology subtyping of glialneuronal tumors

[jaclyn-taroni]

⚠️ Review (& merge) #1014 first ⚠️

Purpose/implementation Section

What scientific question is your analysis addressing?

Subtyping glialneuronal tumors per the WHO 2016 CNS guidelines & instructions on #996

What was your approach?

Here I'm adding a notebook to the molecular-subtyping-pathology module to do the subtyping. I'm using the exclusion criteria for LGAT subtyping (e.g., the JSON file from that module being altered in #1014) as inclusion criteria in this notebook.

What GitHub issue does your pull request address?

Closes #996

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

Which areas should receive a particularly close look?

The output – analyses/molecular-subtyping-pathology/results/glialneuronal_tumor_subtypes.tsv – should be carefully reviewed. As should the logic in analyses/molecular-subtyping-pathology/clinical-subtyping-glialneuronal-tumors.Rmd.

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.

[kgaonkar6]

LGTM, I ran the code and checked the output files and it satisfies the expected updates requested from from #996

[jharenza]

Hi @jaclyn-taroni ! The logic here looks perfect to me, but I have some requests and one request and one question.

I request to rename clinical-subtyping-glialneuronal-tumors.nb* to pathology-harmonized-diagnosis-glialneuronal-tumors.nb* since this data is derived from pathology reports and is going to become harmonized diagnoses.

My question here is similar to that on my review of #1016 - do we want to make these separate scripts or add to the 03-incorporate-pathology-feedback.nb.Rmd script? (I know 03 is quite long, so I would be ok with adding separate scripts, and for samples subtyped, they should be added to compiled_molecular_subtypes_with_clinical_pathology_feedback.tsv.

The case for these is a bit different - we will actually remove these from the subtyping file since they do not get subtyped, so I think we should add that step here.

[jaclyn-taroni]

My question here is similar to that on my review of #1016 - do we want to make these separate scripts or add to the 03-incorporate-pathology-feedback.nb.Rmd script? (I know 03 is quite long, so I would be ok with adding separate scripts, and for samples subtyped, they should be added to compiled_molecular_subtypes_with_clinical_pathology_feedback.tsv.

The case for these is a bit different - we will actually remove these from the subtyping file since they do not get subtyped, so I think we should add that step here.

@jharenza I don't follow - should I be adding something to the notebook I'm adding here?

I will rename the notebook. As I said in #1016 (comment), we should prioritize making sure the logic is correct and then compile the results and reorganize the module as necessary.

[kgaonkar6]

Suggested change

	histologies_file <- file.path(data_dir, "pbta-histologies.tsv")
	histologies_file <- file.path(data_dir, "pbta-histologies-base.tsv")

I believe the scripts here should use pbta-histologies-base.tsv since the pbta-histologies.tsv is only created later in molecular-subtyping-integrate module.

[jaclyn-taroni]

I think that's probably true here but not in #1016 (https://github.com/AlexsLemonade/OpenPBTA-analysis/pull/1016/files#r625098165) since that's downstream of CRANIO subtyping... does that seem correct to you?

[jaclyn-taroni]

Addressed in da8ae7b

[jharenza]

This logic now looks good to me and the N (14) is correct.

[jharenza]

@jaclyn-taroni we should also add tumor_descriptor here as well.

[jaclyn-taroni]

@jaclyn-taroni we should also add tumor_descriptor here as well.

Done! When I ran this with v19 data, 7316-2285 was removed. This seems to be expected/desired because the pathology_diagnosis associated with this sample ID has been updated to Glial-neuronal tumor NOS but wanted to confirm with you @jharenza before merging.

[jharenza]

Done! When I ran this with v19 data, 7316-2285 was removed. This seems to be expected/desired because the pathology_diagnosis associated with this sample ID has been updated to Glial-neuronal tumor NOS but wanted to confirm with you @jharenza before merging.

yes, this looks expected!