GroupReadsByUmi issue

[liuhepkk]

Dear engineer：

There are some fusion templates which R1 R2 may be from different chr in pair end reads.
question1:
Are these templates can be passed after GroupReadsByUmi ?
I was confused of the docs:
" Reads are aggressively filtered out so that only high quality read pairs with both ends mapped are taken forward."

question2:
It is that mean only the templates which R1 R2 from the same chr can be passed ？ Will the fusion analysis results can be influenced ?

Thanks.
Look forward to your favourable reply!

[nh13]

The support for grouping R1 and R2 across contig was added here: #648

I don’t think we’ve done a release with it, so you’ll need to build off master.