Compile pathology_free_text_diagnosis subtypes

[kgaonkar6]

Purpose/implementation Section

What scientific question is your analysis addressing?

Adding back the individual pathology_free_text_diagnosis subtypes to the compiled_molecular_subtypes.tsv

What was your approach?

There are 2 kinds of updates: The samples in the files below have molecular-subtyping results which are already part of the compile file analyses/molecular-subtyping-pathology/results/compiled_molecular_subtypes.tsv so we will be updating the values for these samples using outputs from the pathology-free-text-diagnosis terms. File	Logic to include as results	Previous subtype	Description ticket
analyses/molecular-subtyping-pathology/results/lgat-pathology-free-text-subtypes.tsv #1060	Update existing values	LGAT	#1000
analyses/molecular-subtyping-pathology/results/cranio_adam_subtypes.tsv #823	Update existing values	CRANIO	#994
analyses/molecular-subtyping-pathology/results/glialneuronal_tumor_subtypes.tsv	Update existing values	LGAT	#996

The following subtypes can be directly added to analyses/molecular-subtyping-pathology/results/compiled_molecular_subtypes.tsv as well since these subtypes don't have molecular-subtyping results File	Logic to include as results	Description ticket
analyses/molecular-subtyping-pathology/results/cns-lymphoma-subtypes.tsv	Append	#1057
analyses/molecular-subtyping-pathology/results/juvenile-xanthogranuloma-subtypes.tsv	Append	#1056
analyses/molecular-subtyping-pathology/results/choroid_plexus_papilloma_subtypes.tsv	Append	#1065
analyses/molecular-subtyping-pathology/results/meningioma_subtypes.tsv	Append	#1013

What GitHub issue does your pull request address?

1061

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

Which areas should receive a particularly close look?

• I added just 1 script here to compile the pathology_free_text_diagnosis here since even though there are 2 types of updates they are all from pathology_free_text_diagnosis. Let me know if you would prefer it to be divided into 2 scripts.
• 04-incorporate-pathology-feedback.Rmd needs some careful review
  • like PT_00G007DM was coded to ETMR, C19MC-altered in but is not needed anymore since there are no duplicate subtypes for this patient in v19
  • PT_EHE800JJ is coded in 04-incorporate-pathology-feedback.Rmd but is removed in v19

Is there anything that you want to discuss further?

• There are 2 subtypes for this medulo BS_5BMNK8SY , it seems to be from matching to multiple RNAseq #1085
• In addition the compiling and adding pathology_free_text_diagnosis subtypes there are other updates to compiled_molecular_subtypes.tsv from the hotspots updates / downstream molecular-subtype modules

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)?

tables

What is your summary of the results?

Reproducibility Checklist

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

Documentation Checklist

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

[kgaonkar6]

The following issues were fixed:

• Removed integrated_diagnosis terms in LGG,To be classified pathology_free_text_diagnosis updates should only have harmonized_diagnosis
• Glial and choroid plexus needed Notes to be added
• Only add harmonized_diagnosis for subtypes from
  • pathology_diagnosis updates only ( neurocytoma )
  • pathology_feedback only (chordoma )
• Duplicate for BS_5BMNK8SY fixed
• PT_EHE800JJ is removed from 04-incorporate-pathology-feedback.Rmd because it is removed as per v19

[jharenza]

Could you also pop #1108 into this PR since it is in the same module?

https://github.com/AlexsLemonade/OpenPBTA-analysis/blob/fad3359c16e7dbcff5169d08f854769c55f57821/analyses/molecular-subtyping-pathology/01-compile-subtyping-results.Rmd#L283-L285

should be

  mutate(integrated_diagnosis = case_when(molecular_subtype == "CRANIO, ADAM" ~"Adamantinomatous craniopharyngioma",
                                          molecular_subtype == "CRANIO, PAP" ~"Papillary craniopharyngioma", 
                                          TRUE ~ NA_character_),

[jharenza]

I also wanted to check that you plan to update molecular-subtype-integrate after this PR with updated harmonized_diagnosis since I do not see those filled in here... in which case, I think we may need a ticket for that one?

[kgaonkar6]

In the latest commit, I've updated the typo Adamantinomatous craniopharyngioma and added back meningioma.

About molecular-subtype-integrate didn't we decide that we will only run the module per data release so that the pbta-histologies.tsv file matches? Maybe I mis-understood, a ticket will be good.

[jharenza]

That's possible, but I think the way it is set up now is via an excel sheet matching, right, and there needs to be some logic added for updating the harm dx based on free text dx, rather than just pulling over path dx, right?

[kgaonkar6]

That's possible, but I think the way it is set up now is via an excel sheet matching, right, and there needs to be some logic added for updating the harm dx based on free text dx, rather than just pulling over path dx, right?

Hmm I don't think the logic needs to be updated in the integrate step.

Any BS_ID already subtyped ( ie. exists in compiled_molecular_subtypes_with_clinical_pathology_feedback_and_report_info.tsv) is not going to be part of the code that pulls harm_dx/broad_hist/short_hist from the excel sheet. That part is only for the non-subtyped + pathology_diagnosis=="Other" BS_IDs. We could do have molecular-subtyping-integrate rerun PR to see if that's true or I could have missed some key logic.

[jharenza]

gotcha! that makes sense.