Proposed Analysis: Fusion files specifically for consumption by molecular subtyping analyses

[jaclyn-taroni]

Related to this thread: #315 (comment)

What are the goals of the analysis?

To generate files that contain information about the presence or absence of specific fusions or genes participating in fusions to be used in generating subtype labels for:

• Ependymoma tumors Proposed Analysis: Molecularly subtype ependymoma tumors #245
• Non-MB/ATRT embryonal tumors Proposed Analysis: Molecularly subtype non-MB/ATRT embryonal tumors #251

What input data are required for this analysis?

We can use the putative oncogenic fusion file for this: pbta-fusion-putative-oncogenic.tsv

What is the expected output?

I think output here will be most useful if it is divided into two files in tabular format: 1) fusions relevant for ependymoma tumors and 2) non-MB/ATRT embryonal tumors.

Below I include what I think the files should contained based on my understanding of #245 and #251, where columns with a name that consists of a single gene will indicate if that gene is included in any fusion event.

Ependymoma

Toy example:

Kids_First_Biospecimen_ID	RELA	C11orf95--YAP1	YAP1--MAMLD1	YAP1--MAMLD2	YAP1--FAM118B	C11orf95--MAML2	PTEN--TAS2R1	C11orf95--RELA	LTBP3--RELA
BS_XXXXXXXX	1	0	0	0	0	0	1	1	1

Non-MB/ATRT embryonal tumors

#251 states:

[ETMR, C19MC-altered] tumors have focal amplification of the miRNA cluster on chr19 (denoted C19MC) and often have gene fusions involving TTYH1 and chr19 miRNA cluster genes.

Side note, I think this is the relevant publication for that: Kleinman et al. Nature Genetics. 2014.

Because it seems that this cluster would contain over 40 gene symbols (Wikipedia), I propose listing which, if any, fusions with TTYH1 are present in a sample.

Here's a toy example:

Kids_First_Biospecimen_ID	TTYH1_fusions	MN1	MN1--BEND2	MN1--CXXC5	FOXR2	CIC-NUTM1
BS_XXXXXXXX	TTYH1--<Gene X>, <Gene Y>--TTYH1	1	0	1	0	0
BS_XXXXXXXX	NA	0	0	0	1	0

I don't understand from #251 that the orientation of the MN1--BEND2 and MN1--CXXC5 fusions matters, so I believe BEND2--MN1 and CXXC5--MN1 count.

How long do you expect is needed to complete the analysis? Will it be a multi-step analysis?

1-2 days

Who will complete the analysis (please add a GitHub handle here if relevant)?

Unassigned. Whoever picks this up, please double check the molecular subtyping tickets and ensure that you agree with my summary!

What relevant scientific literature relates to this analysis?

See subtyping tickets.

[dmiller15]

I can take this.

[jaclyn-taroni]

Hi @dmiller15, thank you and welcome to the project! Please see our contributing guidelines and the Implementing an Analysis and How to Add an Analysis sections of the README for more information before you implement this analysis.

You may also find this documentation about the putative oncogenic fusion file helpful: https://github.com/AlexsLemonade/OpenPBTA-analysis/blob/master/doc/data-formats.md#derived-fusion-files

If you have any questions, please let us know! Thanks again!

[jaclyn-taroni]

Closed via #410