Updated analysis: Ependymoma subyping - add spinal EPN type

[jharenza]

What analysis module should be updated and why?

https://github.com/PediatricOpenTargets/OpenPedCan-analysis/tree/dev/analyses/molecular-subtyping-EPN

Add subtypes:

• EPN, SP (EPN, spinal)
• EPN, SP-MYCN (EPN, spinal-MYCN amplified)

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

We should update the https://github.com/PediatricOpenTargets/OpenPedCan-analysis/blob/dev/analyses/molecular-subtyping-EPN/results/EPN_all_data.tsv to add spinal into the diease_group and add additional alteration columns for MYCN amplification, chr22, NF2. K28M will be added in #424

Criteria:
EPN, SP

• CNS_region == spine and
• No MYCN amplification and
• Frequent loss of Chr22 and mutations in NF2 can occur in this subtype

EPN, SP-MYCN

• CNS_region == spine and
• MYCN amplification
  Note: H3 K28M can also rarely occur with MYCN amplification in this subtype

Relevant literature: https://academic.oup.com/neuro-oncology/article/23/8/1231/6311214

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

CNVs: consensus_wgs_plus_cnvkit_wxs.tsv.gz
Chr arms: broad_values_by_arm.txt
cns region: histologies.tsv
mutations: snv-consensus-plus-hotspots.maf.tsv.gz

Note: use data.table::fread() to load the maf

When do you expect the revised analysis will be completed?

1 week

Who will complete the updated analysis?

@rjcorb

[sickler-alex]

For SP, chr22 and NF2 aren't required? The other EPN subtypes have different levels of priority with ST RELA having a higher priority than ST YAP1 then PF A and finally PF B. What priority should be given to these new variants? Basically, if a sample has two different subtypes only the higher priority one is given. This can be done for any subtype, but is only being done for EPN. If a sample has multiple subtypes with the same priority all of those are kept.

[jharenza]

For SP, chr22 and NF2 aren't required?

Correct, but are often associated so it helps with the confidence

The other EPN subtypes have different levels of priority with ST RELA having a higher priority than ST YAP1 then PF A and finally PF B. What priority should be given to these new variants?

I'm not sure we need a priority - if it has one of the fusions, it should only land in that subtype, but maybe that's what you're asking

These should not fit into the other categories at all if they are spinal location, so we can make an option for location of spine removing them from those types- would that work?

[sickler-alex]

For SP, chr22 and NF2 aren't required?

Correct, but are often associated so it helps with the confidence

The other EPN subtypes have different levels of priority with ST RELA having a higher priority than ST YAP1 then PF A and finally PF B. What priority should be given to these new variants?

I'm not sure we need a priority - if it has one of the fusions, it should only land in that subtype, but maybe that's what you're asking

That's what I'm asking. So I can add this as priority 1 that way we'll remove the other subtypes if they were assigned.

These should not fit into the other categories at all if they are spinal location, so we can make an option for location of spine removing them from those types- would that work?