I obtained the GTF file from Cirquant, and from the cookbook, I understand the meaning of each column. I have a question about the "score" column, which represents "CPM of circRNAs (#BSJ / #Mapped reads)." How can this score be greater than 1? For example, I received a score of 763.5227, which seems quite high.
Additionally, I noticed that many circRNA coord_ids have multiple BSJ (Back Splicing Junctions), while the FSJ (Forward Splicing Junctions) is 0. How can I explain this situation? Do these regions consist solely of circular transcripts?
Finally, the bsj obtained from some gtf files has too large a difference, with the highest being tens of thousands and the lowest being dozens or even 0. This makes it impossible to perform a difference analysis on a large sample. I don't know if you have observed this. What is the reason for this, and is there any improvement? Or how should I deal with this?
I obtained the GTF file from Cirquant, and from the cookbook, I understand the meaning of each column. I have a question about the "score" column, which represents "CPM of circRNAs (#BSJ / #Mapped reads)." How can this score be greater than 1? For example, I received a score of 763.5227, which seems quite high.
Additionally, I noticed that many circRNA coord_ids have multiple BSJ (Back Splicing Junctions), while the FSJ (Forward Splicing Junctions) is 0. How can I explain this situation? Do these regions consist solely of circular transcripts?
Finally, the bsj obtained from some gtf files has too large a difference, with the highest being tens of thousands and the lowest being dozens or even 0. This makes it impossible to perform a difference analysis on a large sample. I don't know if you have observed this. What is the reason for this, and is there any improvement? Or how should I deal with this?