Add pnoc003 clinical review

[kgaonkar6]

Purpose/implementation Section

What scientific question is your analysis addressing?

As part of molecular-subtype-HGG analysis we assign a HGG or DMG subtype from looking for K28M histone variants, in this PR we wanted to identifying samples which have differences between the clinically reviewed subtype and subtype from molecular-subtype-HGG module in PNOC003 and update those to the clinical reviewed subtypes.

What was your approach?

The table from the original issue added by @jharenza is added as file in input folder. I used the file to check for changes and updated 3 WXS and 3 RNA-Seq samples from PT_NK8A49X5, PT_QA9WJ679, PT_WGVEF96B by matching to the values in the reviewed files.

As discussed in #798 , I'm moving the code from molecular-subtyping-HGG to molecular-subtyping-pathology

• as a separate notebook before incorporating path feedback

This order was decided after discussing with @jharenza, since the pathology feedbacks are major changes to the diagnosis , so once the clinical feedbacks are incorporated (as step 2) the subtype/diagnosis would need additional review from pathology as a final step in this module ( as step 3).

What GitHub issue does your pull request address?

#751

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

Which areas should receive a particularly close look?

I've created the file for v17 (which doesn't have subtypes for cell-lines) so I believe that will need to be updated once #797

Is there anything that you want to discuss further?

I've coded the checks specifically to only substitute and not assign H3 wildtype to DMG, H3 K28 if clinically the subtype was K28M mutant
https://github.com/kgaonkar6/OpenPBTA-analysis/blob/6e70a6795719007cd0233074b3fff7218d84bbb3/analyses/molecular-subtyping-HGG/10-pnoc003-clinical-review-update.Rmd#L145-L155

I did that because some subtypes have additional subtype like HGG IDH or BRAF V600 that need to be retained

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)?

table

What is your summary of the results?

results/compiled_molecular_subtypes_with_pathology_feedback_clinical_review.tsv

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.

[jaclyn-taroni]

Thanks for moving this to this module @kgaonkar6 ! I think the functionality is where it needs to be. I had some questions about tying this to a specific release and I also had a thought about how to make some of the logic more readable.

[jaclyn-taroni]

As I'm looking at this, I am wondering if, for the purposes of this notebook, we might want to standardize the labels between the H3 Status and molecular_subtype so you can essentially ask "are these the same?" That logic might be more straightforward. Thoughts? We'd probably also want to filter out the no reports - I think this is essence what this is accomplishing as well.

[jaclyn-taroni]

I suppose this extends to the Notes step, too.

[kgaonkar6]

I thought about changing the values in H3 Status to what we have in molecular_subtype, but since we have some additional values from IDH/BRAF alterations I assumed that it would be the same logic of grepl because of the IDH/BRAF alterations subtyping that is not part of clinical review. For example these samples have the additional info from molecular-subtyping-HGG :

Kids_First_Participant_ID	sample_id	Kids_First_Biospecimen_ID	molecular_subtype	integrated_diagnosis	short_histology	broad_histology	Notes	tumor_descriptor
PT_0MXPTTM3	7316-3220	BS_8SYN7GXG	HGG, IDH	High-grade glioma	HGAT	Diffuse astrocytic and oligodendroglial tumor	Updated via OpenPBTA subtyping	Initial CNS Tumor
PT_0MXPTTM3	A09410	BS_R7NTZR4C	HGG, IDH	High-grade glioma	HGAT	Diffuse astrocytic and oligodendroglial tumor	Updated via OpenPBTA subtyping	Initial CNS Tumor
PT_0MXPTTM3	A09410	BS_H3TYF6GD	HGG, IDH	High-grade glioma	HGAT	Diffuse astrocytic and oligodendroglial tumor	Updated via OpenPBTA subtyping	Initial CNS Tumor
PT_HGM20MW7	7316-3221	BS_FK3B5SDH	DMG, H3 K28, BRAF V600E	Diffuse midline glioma	HGAT	Diffuse astrocytic and oligodendroglial tumor	Updated via OpenPBTA subtyping	Initial CNS Tumor
PT_HGM20MW7	A09446	BS_5VPM0F36	DMG, H3 K28, BRAF V600E	Diffuse midline glioma	HGAT	Diffuse astrocytic and oligodendroglial tumor	Updated via OpenPBTA subtyping	Initial CNS Tumor
PT_HGM20MW7	A09446	BS_52ETE050	DMG, H3 K28, BRAF V600E	Diffuse midline glioma	HGAT	Diffuse astrocytic and oligodendroglial tumor	Updated via OpenPBTA subtyping	Initial CNS Tumor

Maybe these IDH, BRAF alterations should also be clinical reviewed so we can match the values?

[jaclyn-taroni]

I see - so in the PT_0MXPTTM3 case, H3 status is wildtype which agrees with the molecular subtyping status and in the PT_HGM20MW7 H3 status and the subtyping is in agreement, it's just the matter of BRAF V600E also being included.

Taking a step back - my understanding of this addition is that we're essentially looking for any K28M mutations that were missed because they were not detected in the SNV consensus for whatever reason (from #751 and some additional context in #735). I had some ideas about how we might approach this differently for clarity reasons, but I think we can go with this solution and have you add comments throughout at each condition within calls to case_when() (lines 180-197).

[jaclyn-taroni]

This looks good to me! Do we need @jharenza to take a look before merging?

[jharenza]

Looks good, thanks @kgaonkar6 !