Is the classic WGS data well compatible with circle-seq

[zhang616123]

Dear, I recently read your excellent work on circle-map, thank you for your contribution, but I have some small confusion, I noticed that in the middle part of the paper you used data from Circle-seq to test circle-map , This method enriches eccDNA before sequencing. Is this necessary for circle-map? If not, can the data generated by the classic WGS sequencing process be used? If possible, 1) Will noise from other genomic regions have an adverse effect; 2) Do we need to adopt more relaxed screening criteria for the output of circle-map? Best.

[iprada]

Dear @zhang616123,

Thanks for you nice words regarding our work.

Regarding your questions:

I think that Circle-Map could work very well in standard WGS data, as long as you apply strong downstream filters to remove noise. You could, of course, validate the results experimentally. You will probably detect less circles than in non-circle-enriched data. But whether this will sufficient depends on your research question.

If you try it, I would like to hear the results. Feel free to send me an email if it works :)

[zhang919]

Dear @iprada, Thanks for your attention, after read your reply, I am trying apply Circle-map on standard WGS data. Fortunately, It works better than expected, I got average 10000+ candidates eccDNAs，but interesting once it can pass the screening, the number of eccDNA is always very small. for example if I apply split reads > 2 to filter. Could you please give me some suggestions about it, Thank you very much. Best. Zhang

[iprada]

Hi @zhang919 ,

How many circles do you get after the >2 split read filter?

Best,

Inigo

[zhang919]

Hi @iprada I'm sorry to ignore this point. I got about 100-200 circles from the split reads > 2 filter.

[iprada]

Hi,

As you have non-enriched data, I do not find surprising that you are on the 100-200 circle range.

Best,

Inigo