Suggestions for Transcript Reconstruction from Novel MAJIQ Junctions

[Phillip West]

Hi,

I’m currently using MAJIQ to detect local splicing variations, including novel junctions and cryptic exons, in a disease-focused mouse RNA-seq dataset.

Do the authors of MAJIQ or community members have any recommended approaches, tools, or best practices for transcript-level reconstruction from MAJIQ outputs? Specifically, I’m looking for a way to "stitch" novel junctions into transcript models, ideally in a way that maintains compatibility with existing annotations (Ensembl GRCm39). I'm looking to generate the predicted nucleotide sequences for the novel splice changes.

Cheers, Phillip

[bsl...@seas.upenn.edu]

Dear Phillip,

MAJIQ does not do transcript-level reconstruction/quantifications. Furthermore, I would not advise attempting to recover transcript-level quantifications from MAJIQ outputs. Given novel junction coordinates, I do think it's possible to augment an existing transcriptome annotation with some new possible transcripts, but MAJIQ does not include tools to do that.

Best Regards,

Barry