m6A Detection for Novel Fusions

[hannalee809]

Hi! First of all, thank you for developing m6anet. It has been a crucial tool in my project, and I truly appreciate the effort put into creating such a valuable resource.

I am currently exploring ways to apply m6anet for m6A detection in novel fusion transcripts identified using the JAFFA pipeline . Since JAFFA provides both fusion annotations and corresponding FASTA sequences, I was wondering if you have any recommendations on how to adapt m6anet for detecting m6A modifications in these novel fusion transcripts.

Would it be feasible to incorporate these sequences into a reference transcriptome for alignment, or do you have any suggestions on the best approach for analyzing m6A modifications in fusion transcripts? If there are any existing workflows or adaptations that might be useful, I would greatly appreciate any insights.

Looking forwards to you thoughts, thank you!

[rugilemat]

I also had a very similar question - would it be possible to run m6anet using a reference transcriptome which includes identified novel isoforms, e.g. after analysis with bambu or similar tools, followed by SQANTI3 QC, i.e. using SQANTI fasta and gtf files as a reference?
In this case, should the mapping be done with transcriptome reference rather than genome reference in the first place? Or do you have any other suggestions on how to incorporate novel isoforms, similarly to hannalee809's question?
Thanks!

[yuukiiwa]

Hi @hannalee809 and @rugilemat,

Yes, it is possible to run m6anet with a reference transcriptome with identified new isoforms. Here is what I would do:

1. generate a transcriptome fasta file with the identified new isoforms
2. align the fastq to newly generated transcritome fasta file
3. eventalign with the newly generated transcritome fasta file
4. run m6anet as usual

Thanks!

Best wishes,
Yuk Kei

[rugilemat]

Hi @yuukiiwa, that's great, thanks!

Since SQANTI3 output is a genome fasta, it first needs to be converted to transcriptome fasta for alignment/m6anet run and then the m6anet run can be converted back to genome coordinates after the run?

[yuukiiwa]

Hi @rugilemat,

Yes, you will need to convert the genome fasta from SQANTI3 to transcriptome for running alignment and m6anet. After the m6anet run, then you can convert back to genome coordinates.

Thanks!

Best wishes,
Yuk Kei