Troubleshooting Fusion Detection in Pooled RNA-seq Data

[Sharon Bader]

Hello,

We have cells expressing splicing machinery via lentiviral constructs and have been using STAR-Fusion to detect possible fusions, with a custom reference library that includes the entire human genome plus the inserted sequence.

In our initial pilot experiments, we ran the constructs in parallel and sequenced using bulk RNA-seq. This time, we infected cells individually, pooled the resulting populations, and sequenced them after some period of time. The STAR-Fusion pipeline performed well in the pilot experiment; however, in our current pooled analysis, we're detecting barely any fusions.

Do you have any thoughts on what might be going wrong? Could this be due to an issue with how we constructed the reference library, or might it be related to sequencing depth?

Thanks in advance for your help!

[Brian Haas]

Hi Sharon,

If the fusions are going to be rare events, then you'll want to run STAR-Fusion with the --full_Monty option so it reports everything it finds.  It will likely require some additional filtering on your end to decide on what thresholds you need to employ to differentiate 'adventitious' stuff vs. results related to your experimental design.

Hope this helps,

Brian

--
You received this message because you are subscribed to the Google Groups "Trinity_CTAT_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinity_ctat_us...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/trinity_ctat_users/856e43a7-c758-4878-a999-4416b01d3dd6n%40googlegroups.com.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas