Updated analysis: MB subtyping to use pathology diagnosis

[jharenza]

What analysis module should be updated and why?

https://github.com/AlexsLemonade/OpenPBTA-analysis/tree/master/analyses/molecular-subtyping-MB

What changes need to be made? Please provide enough detail for another participant to make the update.

Currently, this module uses integrated_diagnosis to search for the subset of samples to be subtyped. However, with the rework of rules to determine integrated_diagnosis post molecular subtyping (#748), we should rework this module to search for samples using pathology_diagnosis and the new field for free text pathology diagnosis, which will be present in `pbta-histologies.tsv in #732.

What input data should be used? Which data were used in the version being updated?

pbta-histologies.tsv from V17 instead of from V16

When do you expect the revised analysis will be completed?

unsure

Who will complete the updated analysis?

unsure

[jaclyn-taroni]

@cansavvy since #753 is apparently blocked, why don't we have you plan to pick up this issue? I've filed #786 (HGG) which it may be helpful to have a look at for inspiration re: first look at the pathology_diagnosis and pathology_free_text_diagnosis.

[cansavvy]

Sounds good! I'll take a look at this today!

[cansavvy]

I'll push a small notebook that summarizes this, but I did some poking and MB substyping seems to be a bit more straightforward than some of the other histologies:

• Looks like all samples that have short_histology and integrated_diagnosis as Medulloblastoma also have pathology_diagnosis as Medulloblastoma.
• Most but not all of these samples have medulloblastoma some place in pathology_free_text_diagnosis, but a couple have PNET.
• There are no samples currently classified as short_histology == "Other" that have medulloblastoma as a part of pathology_free_text_diagnosis

[jaclyn-taroni]

Closed via #787