Updated analysis: Update mol subtyping integrate to include ATRT + NBL

[zzgeng]

…r_subtyping_ATRT

Purpose/implementation Section

What scientific question is your analysis addressing?

The NBL and ATRT subtypes need to be added to this so that they show up in the final histology file.

What was your approach?

Add ATRT subtyping results into histology file

What GitHub issue does your pull request address?

d3b-center/ticket-tracker-OPC#504

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

Which areas should receive a particularly close look?

analyses/molecular-subtyping-pathology/01-compile-subtyping-results.Rmd
analyses/molecular-subtyping-integrate/results/histologies.tsv

Is there anything that you want to discuss further?

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

Need to wait until #264 is merged

Results

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

Table

What is your summary of the results?

Add ATRT subtyping results into histologies file.

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]

Can you pull molecular-subtyping-NBL branch and then do a git merge molecular-subtyping-NBL on this branch? Then you will have the NBL files and this can go through the module fully. Then, also change the base branch above via web interface to molecular-subtyping-NBL instead of dev.

[jharenza]

Can you pull molecular-subtyping-NBL branch and then do a git merge molecular-subtyping-NBL on this branch? Then you will have the NBL files and this can go through the module fully. Then, also change the base branch above via web interface to molecular-subtyping-NBL instead of dev.

I have now merged the NBL subtyping PR, so after conflicts are resolved, you can go ahead and merge dev into this

[jharenza]

@ewafula and @zzgeng I made some updates to this 01-compile-subtyping-results.Rmd

• pulled in all possible EPN, ATRT, IHG, and NBL subtypes
• added Notes, short, broad histologies - needed minor updates
• OpenPBTA --> OpenPedCan in Notes and links to modules
• updated some integrated dx terminology to WHO 2021 (CNS embryonal NEC/NOS, EPN, diffuse hemispheric glioma, HGG IDH+H3 wildtype, IHG)
• added integrated dx for ATRT and NBL
• updated function for separate/binding results data frames to account for the methyl + panel data

I will approve, but would be good to have one more set of eyes on analyses/molecular-subtyping-pathology/results/compiled_molecular_subtypes.tsv to confirm all subtypes which required data and not pathology free text have an integrated dx, and they align

[ewafula]

@jharenza, just checked analyses/molecular-subtyping-pathology/results/compiled_molecular_subtypes.tsv
So, chordoma does not require subtying?
Why Neuroblastoma/Ganglioneuroblastoma missing Kids_First_participant_ID and sampled_id?

[jharenza]

@jharenza, just checked analyses/molecular-subtyping-pathology/results/compiled_molecular_subtypes.tsv
So, chordoma does not require subtying?

I was just looking into that, too - I think this is a special case, bc we had to consult with pathology. on these since they are "poorly differentiating" - they may all be manually updated in the clinical and path feedback file. Will check.

Why Neuroblastoma/Ganglioneuroblastoma missing Kids_First_participant_ID and sampled_id?

Let me check this - this module binds with the results from the module, so possible they are missed there, we might need to update, but will check.

[jharenza]

chordoma will be updated in pathology feedback: https://github.com/PediatricOpenTargets/OpenPedCan-analysis/blob/efc08161115c2a53e740a46b9f5630d5d1450fdb/analyses/molecular-subtyping-pathology/results/compiled_molecular_subtypes_with_clinical_pathology_feedback_and_report_info.tsv

Possibly we can rework this later to not be in this module.

[jharenza]

super catch @ewafula !! See NBL results now

[ewafula]

@jharenza, one more question.

All BS_IDs not subtyped (to be classified) don't have integerated_diaginosis (set to NA) as expected. What about these 14 that have molecular_subtype but integerated_diaginosis == NA?

> read_tsv("~/Downloads/compiled_molecular_subtypes.tsv") %>% 
>      filter(is.na(integrated_diagnosis), 
>               str_detect(molecular_subtype, "To be classified", negate = TRUE), 
>               str_detect(molecular_subtype, "to be classified", negate = TRUE)) %>% 
>     select(Kids_First_Biospecimen_ID, molecular_subtype, integrated_diagnosis) %>% 
>     print(n = 14)
# A tibble: 14 x 3
   Kids_First_Biospecimen_ID molecular_subtype            integrated_diagnosis
   <chr>                     <chr>                        <chr>               
 1 BS_M7HCF7DW               IHG, ROS1-altered, TP53 loss NA                  
 2 BS_M28HG0NT               IHG, ROS1-altered, TP53 loss NA                  
 3 BS_CXVGG342               IHG, ROS1-altered, TP53 loss NA                  
 4 BS_RG6JK7ZN               IHG, ROS1-altered, TP53 loss NA                  
 5 BS_VH68VD09               EVN                          NA                  
 6 BS_YPFQ9ETM               EVN                          NA                  
 7 BS_BMHYK93N               CNC                          NA                  
 8 BS_TGW6HC4X               CNC                          NA                  
 9 BS_DGTPWJ59               CNC                          NA                  
10 BS_FS3DD2PH               CNC                          NA                  
11 BS_SNC5WMVN               CNC                          NA                  
12 BS_6SA5B35V               EVN                          NA                  
13 BS_8F9BDWY4               EVN                          NA                  
14 BS_P26GF314               EVN                          NA                  
>

Not a big deal - should this be kept consistent in the molecular_subtype column for BS_IDs with not subtypes:

• *, To be classified (* == all other dx)
• NBL, to be classified

Maybe, mutate NBL, to be classified to NBL, To be classified to be consistent. Alternatively, we need to update in molecular-subtype-NBL module output.

[jharenza]

1 BS_M7HCF7DW IHG, ROS1-altered, TP53 loss NA
2 BS_M28HG0NT IHG, ROS1-altered, TP53 loss NA
3 BS_CXVGG342 IHG, ROS1-altered, TP53 loss NA
4 BS_RG6JK7ZN IHG, ROS1-altered, TP53 loss NA

For these, I updated - missed this bc of the TP53 status so I updated the code to just grepl the alteration determining the subtype for all of HGG.

For the neurocytoma ones, they will not get an integrated dx, only a harm dx - since the subtype is not based on molecular data, just the primary site and free text.

Ohh good catch again about to be classified. Let me change that in the NBL module - it is important for other things which escape me now!

[jharenza]

@ewafula ready for re-review

[ewafula]

@jharenza, this now looks ok to me. Thank you for updating.

Approving