Fusion gene analysis?

[apsteinberg]

Hi there,

Thank you again for creating this wonderful tool! I had a question regarding your fusion gene analysis in your recent biorxiv paper: https://www.biorxiv.org/content/10.1101/2021.04.21.440736v1

In the methods, you describe aligning reads to a fusion gene annotation set from JAFFAL, then running bambu on these fusion alignments to discover fusion transcripts. I had a few questions about this:

1. Do you have a pipeline to do this, and if so, would you be willing to share the code? The results you show in Fig 7 of the paper are striking, and I would love to run such an analysis on the datasets I'm working with.
2. For the "fusion gene set" that you describe aligning to in the manuscript, is this just the cDNA fasta file output from JAFFAL? And did you somehow extract reads specific to these regions to perform this alignment or just align the entire fastq to these gene fusions?
3. When bambu was then run on the fusion alignments to discover fusion transcripts, was this done using the "fusion mode"? Further, was the reference gtf just ensembl or was bambu run in denovo mode here?

Thanks for your time and help.

Sincerely, Asher