How does CIRCexplorer2 work with paired-end data?

Question

[IP]

Hi everyone!

I have been using CIRCexplorer2 in single-end RNA sequencing datasets for a while, however, I have never use it for paired end data. In the papers and suplementary methods of circexplorer and circexplorer2 it is clearly explained how does the program identify circRNA in single-end datasets. It aims to identify the backspliced junction read of the circRNA, that's it, a read that maps in a non-colinear order to the genome.

However, for paired end datasets, using tophat-tophat-fusion, I think that there could be different aproaches to identify circRNA, as one could have, different indicators for circRNAs, for example:

• The left reads with the backsplicing read and the right read mapping to an exon.
• The right part of the read will map in the genome before the left part

Does anyone know how does CIRCexplorer2 adress this problems?

Thanks for reading,

Answer 1

[kepbod]

The latest version of CIRCexplorer2 has supported paired end datasets. See http://circexplorer2.readthedocs.io/en/latest/tutorial/alignment/ for details.

[IP]

Yeah, I know (I have use it and I got very interesting results). My question is how does it work theoretically with paired-end data. Does it aim to identify only backspliced junction reads? Or does it take another kind of information into account? like in this case:

The right part of the read will map in the genome before the left part