Update EFO MONDO file with EFO IDs

[afarrel]

What data file(s) does this issue pertain to?

efo-mondo-map.tsv

What release are you using?

v8/v9

Put your question or report your issue here.

Update EFO/MONDO designations in efo-mondo-map.tsv
Assign all EFO IDs based on information from Open Targets and OLS. (EFO ID may start with "MONDO_######", etc)

Who will work on this:

@sangeetashukla and @afarrel

[sangeetashukla]

Hi @jharenza @afarrel

I reviewed the EFO and MONDO codes mapping, and found some cancer_groups where the codes may need to be updated. I have attached the file here, along with comments.

EFO_MONDO_mapping_review.txt

[afarrel]

Hi @sangeetashukla ... The format of the file seems weird. Can you fix it into TSV format. Similar to the format in the previous releases?

Thanks

[jharenza]

Hi @sangeetashukla ... The format of the file seems weird. Can you fix it into TSV format. Similar to the format in the previous releases?

Thanks

For some reason, GitHub doesn't allow tsv files to be attached, but @sangeetashukla maybe we should create a pr to add an efo-mondo-map module so it's easier to review for future iterations?

[afarrel]

Hi @sangeetashukla ... The format of the file seems weird. Can you fix it into TSV format. Similar to the format in the previous releases?
Thanks

For some reason, GitHub doesn't allow tsv files to be attached, but @sangeetashukla maybe we should create a pr to add an efo-mondo-map module so it's easier to review for future iterations?

Yea - but you can it allows tab-delimited txt files? Or no?

[jharenza]

Yea - but you can it allows tab-delimited txt files? Or no?

Yes! Just also thinking ahead also as this will change many times that we may want to have some record in GitHub with tracking notes for changes aside from tickets?

[sangeetashukla]

Yea - but you can it allows tab-delimited txt files? Or no?

Yes! Just also thinking ahead also as this will change many times that we may want to have some record in GitHub with tracking notes for changes aside from tickets?

Sure, here it is

[sangeetashukla]

Hi @sangeetashukla ... The format of the file seems weird. Can you fix it into TSV format. Similar to the format in the previous releases?
Thanks

For some reason, GitHub doesn't allow tsv files to be attached, but @sangeetashukla maybe we should create a pr to add an efo-mondo-map module so it's easier to review for future iterations?

Yea - but you can it allows tab-delimited txt files? Or no?

@afarrel
I just realized my file was still a .txt although I pushed it via a PR too.
Attaching it here, and will also convert it to a tsv and push to the PR again.

EFO_MONDO_mapping_review.txt

[afarrel]

Some revisions are needed. A careful look at OLS and platform.opentargets.org will help us find the appropriate "EFO ID"
Will Discuss offline with @sangeetashukla.

[afarrel]

efo-mondo-map_NEW.xlsx

Hi @sangeetashukla and @jharenza
I took at look at the list Sangeeta put together and the original list and looked up the EFO codes from OLS and the OpenTargets platform. The 'NEW' codes are in Column E Some are missing with notes and I have specific questions below so we can decide and confirm which codes we should use:

For "Germinoma;Teratoma" Germinoma is EFO_0000514, Teratoma is MONDO_0002601. There is only one patient with this 'combination/histology' - which means they will be excluded from the analysis modules the require n >= 5.

For "Choroid plexus cyst" - Are we including cysts even though it's not a tumor? Also, N = 1 patient with 1 NGS and 1 RNAseq which means they will be excluded from the analysis modules the require n >= 5.

Confirm CNS Embryonal tumor:MONDO_0018843 (Actual Description is: embryonal carcinoma of the central nervous system)

"CNS neuroblastoma" - will have the same EFO code as neuroblastoma. - The EFO code covers Neuroblastoma and "Central Neuroblastoma" as well as other subtypes of neuroblastoma? Would we need to get subtype EFO/MONDO codes for NBL too?

"Diffuse midline glioma" - (DMG) has no code. Currently using DIPG code. DMG with H3 K27M-Mutant does have NCIT codes: NCIT_C129309; DOID_0080684

"Dysplasia/Gliosis;Glial-neuronal tumor NOS" - Should there be separates histologies? Gliosis is HP_0002171 and mixed neuronal-glial tumor is MONDO_0016729.

"Epilepsy" - Should we include this: Epilepsy isn’t a tumor; N = 2 1 WGS and 1 RNAseq - same patient

"Extraventricular Neurocytoma" - 1 patient with 1 WGS and 1 RNAseq - The old EFO code (EFO_0007201) needs to be reviewed.

"High-grade glioma/astrocytoma" - high-grade glioma: MONDO_0100342; astrocytoma: EFO_0000272 (synonym - astrocytic tumor); high grade astrocytic tumor: MONDO_0016680 - Which is most appropriate?

"Low-grade glioma/astrocytoma" - Low Grade Glioma:MONDO_0021637 ; Low grade astrocytoma: MONDO_0016685

"Melanocytic tumor" - Confirm we can use the EFO code for 'CNS melanocytic tumor'

"Myxoid spindle cell tumor" - myxoid isn't specified - maybe need to request subtype in future?

[jharenza]

closed with d3b-center/OpenPedCan-analysis#99 merged