Result analysis about sqanti2

[Magdoll]

Hi @ljw90607 ,

If most of your data is counted as anti-sense, most likely it's because the data is not stranded.

SQANTI2 trusts the input sequence as the sense strand. By default minimap2 is called with -uf parameter. If your input data is not stranded, you will either need to manually modify the minimap2 parameter in sqanti_qc2.py or modify the input sequences so they are stranded.

--Liz

[ljw90607]

Dear @Magdoll,

We have tried to use the sqanti_qc2.py without the option -uf which then would be mapping the transcripts as unknown. But we found out that even without the -uf parameter, I was still getting the half of my reads as antisense.
Would there be any suggestion on how I can annotate these antisense strands correctly?
Thank you for your help.

[Magdoll]

Hi @ljw90607 ,
Since I'm not familiar with nanopore data, I will not be able to help much further. I can only offer the experience for PacBio Iso-Seq data.

Thanks,
-Liz