Clarification on strand-related categories in read-Level Analysis

[youyupei]

Hi SQANTI team,

Thanks for developing this tool. I wanted to get some clarification on the output of SQANTI.

I ran SQANTI on my long-read data using the following command:

sqanti_qc3.py {fastq (unstranded cDNA reads)}  {genecode_annotation.gtf} {reference_genome.fa} --fasta --skipORF --force_id_ignore -d {output} --report html

The run was completed successfully, and each read was assigned a structural category. However, I’m uncertain about how to interpret some of these categories, particularly those related to strand information, such as “Antisense” and “Intergenic.” Given that my input reads we unstranded cDNA, where both strands of the RNA could be sequenced, would this affect the proportion of “Antisense” reads I observe?

Additionally, I would appreciate any comments on the validity of running SQANTI3 at the read level.

Thank you in advance for your help.