Reciprocal and kinase

[kgaonkar6]

Purpose/implementation Section

What scientific question is your analysis addressing?

Add kinase domain retention status for fused genes since this information will be needed to filter BRAF and other kinase gene fusions that we use for LGAT subtyping. We also want to check if the fusion is a reciprocal that is if the fusion callers called GeneX--GeneY and GeneY--GeneX.

What was your approach?

First, I added the LeftBreakpoint and RightBreakpoint column since we need this information to annotate domain retention. (Earlier ,we had removed these columns so that 1 unique fusion row per Sample could be retained)

Then, we will be using fusion_driver function from annoFuse to add kinase domain status per Gene1A (5 Gene) and Gene1B (3 Gene) in columns
DomainRetainedGene1A and DomainRetainedGene1B.For each kinase gene the Domain retention annotation will be as follows

Domain retention annotation	Description
DomianRetainedGene1A == Yes	LeftBreakpoint downstream of domain end in any fusion
DomianRetainedGene1A == Partial	LeftBreakpoint within domain boundaries in any fusion
DomianRetainedGene1A == No	LeftBreakpoint upstream of domain start in any fusion
DomianRetainedGene1B == Yes	RightBreakpoint upstream of domain start in in-frame fusion
DomianRetainedGene1B == Partial	RightBreakpoint within domain boundaries in in-frame fusion
DomianRetainedGene1B == No	RightBreakpoint downstream of domain end in any fusion

Within the function the base function pfam domain annotation annotates the retention status of domains per breakpoint and domain ID & Location information from :

Annotation	File	Source
pfamID	http://hgdownload.soe.ucsc.edu/goldenPath/hg38/database/pfamDesc.txt.gz	UCSC pfamID Description database
Domain Location	http://hgdownload.soe.ucsc.edu/goldenPath/hg38/database/ucscGenePfam.txt.gz	UCSC pfamID Description database

For reciprocal status I've added a function to add that information as logical values to a separate column reciprocal_exists. For sample in Sample BS_044XZ8ST we have reciprocal fusion ANTXR1--BRAF and BRAF --ANTXR1 so these fusions will be reciprocal_exists== TRUE

What GitHub issue does your pull request address?

#812

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

Which areas should receive a particularly close look?

I've tried to organize the chunks in 04-project-specific-filtering.Rmd so that there are minimal code changes, please let me know if it is easy enough to follow.

Is there anything that you want to discuss further?

Since I've now added the LeftBreakpoint and RightBreakpoint columns to pbta-fusion-putative-oncogenic.tsv there can be multiple rows per FusionName and Sample if they have multiple breakpoints for the fusion. It doesn't affect the *recurrent-fusion-byhistology.tsv, *recurrent-fused-genes-byhistology.tsv, *recurrent-fusion-bysamplee.tsv and *recurrent-fused-genes-bysampletsv but might affect some other modules that don't unique for FusionName Sample

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?

Kinase domain retention information is added per kinase gene fusion

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]

Are the changes in analyses/fusion_filtering/results/pbta-fusion-recurrent-fusion-bysample.tsv and analyses/fusion_filtering/results/pbta-fusion-recurrently-fused-genes-bysample.tsv expected or unexpected?

[jaclyn-taroni]

This looks good! I had a few questions before I approve.

[jaclyn-taroni]

Can you remind me why we're not looking at the Gene2A and Gene2B here?

[kgaonkar6]

For intergenic fusions, which has Gene1A/Gene2A--Gene1B or Gene1A--Gene1B/Gene2B and similar fusions it get's a little complicated since then we need to check the distance is the same between Gene1A/Gene2A in the reciprocal so I've just stuck to fusions between genes.

[kgaonkar6]

Are the changes in analyses/fusion_filtering/results/pbta-fusion-recurrent-fusion-bysample.tsv and analyses/fusion_filtering/results/pbta-fusion-recurrently-fused-genes-bysample.tsv expected or unexpected?

This is expected because of the different sample selection issues by running sample() when we have multiple samples per Kids_First_Participant_ID.

[jaclyn-taroni]

This is expected because of the different sample selection issues by running sample() when we have multiple samples per Kids_First_Participant_ID.

I would have expected sorting + setting a seed to have prevented that, but there may be some subtlety I'm missing. Either way, beyond the scope of this PR.

[jaclyn-taroni]

👍

[kgaonkar6]

Thanks for the review @jaclyn-taroni !