The question about the boundary of circular RNA

Question

[lxwgcool]

I do some circular RNA researches recently. I have some questions about the boundary of Circular RNA. Since Circular RNA caused by some exons linking back to their previous one, and the RNA detection tool will only report the left position and the right position of the latent circular RNA, is it correct for me to consider the boundary of circular RNA should come from the the boundary of exons?

For example if some of tools report the detected circular rna is (1000, 2000), then i could speculate that both 1000 and 2000 should come from the boundary of some exons.

If my assumption above is incorrect, how about another idea: the majority part of the boundary of circular RNA comes from the boundary of exons. If it is, may i know how many percentage satisfy the second assumption above?

Thanks. I am really appreciate for your helps.

[Brice Sarver]

Are you talking about conventions for linearizing circular molecules? What exactly do you want to know? Are you trying to come up with a convention for analysis?

[lxwgcool]

Thanks so much for your reply. Let me make some detail explanations about my question.

Currently, I have the DNA reference geno (such as chr1.fa), and the DNA pair end reads (such as PE1.fq and PE2.fq). In addition, i already have the RNA annotation file (such as chr1.gtf).

What i want to do is detect the potential circular rna.

I use some tools for detection, such as CIRI and CIRCexplorer. The final report which generated by those tools contain the boundary information of potential circular rna. Here is a part of their report:

--------------------------------->

circRNA_ID chr circRNA_start circRNA_end #junction_reads
chr1:13221|13374 chr1 13221 13374 3
chr1:35245|36073 chr1 35245 36073 44
chr1:89551|90050 chr1 89551 90050 4

<---------------------------------

I think "circRNA_start" and "circRNA_end" should be two boundaries of the corresponding potential circular rna.

For my understanding, circular ran caused by the exon linking back to its previous one. As a result, i think either "circRNA_start" or "circRNA_end" should comes from the boundary of certain exon.

However, i find some of boundaries in the final report of those tools do not belongs to any boundaries of the exon which has been identified in rna annotation file (such has gtf file).

So my question is: are those circular rna correct? Do we just need to pay attention to the circular rna whose boundaries are exactly come from exon.

Thanks so much for your help.