amplicons classification

[GenomeArch]

Hi, Thanks for the great tools you've provided. I've been going through the previous publications "Extrachromosomal DNA is associated with oncogene amplification and poor outcome across multiple cancers" and noticed that ecDNA has been classification has changed from

• Linear
• Heavily rearranged
• Cyclic
• (BFB)
• Linear
  to
• Cyclic
• BFB + Complex-non-cyclic
• Linear
• No amp/Invalid
  Specifically, I'm wondering if the new class "BFB + Complex-non-cyclic" is the same as previous " Heavily rearranged".
  In a more recent paper "Mapping extrachromosomal DNA amplifications during cancer progression" , the author class them as"ecdna", "chramp","noamp", how to compare these classifications？
  If a sample contains multiple positive instances of ecDNA and BFB+, can the sample be classified as ecDNA+?

[jluebeck]

Thanks, this is a great query.

The "BFB + Complex-non-cyclic" class is a composite class that encompasses focal amplification features marked as complex-non-cyclic or BFB. It is not a direct output of AC, but rather a post-processing.

In the Kim, 2020 paper, the differences are that "Heavily rearranged" is now called "Complex non-cyclic" and "Cyclic" is now called "ecDNA". Currently, AC still can report AA amplicons as "Cyclic", however you must check to see if AC marked it as ecDNA-positive or BFB-positive. The "Cyclic" name is a bioinformatic invention to describe the presence of genome cycles, which can be found bioinformatically in both ecDNA or BFB. The Kim, 2024 paper condenses BFB, CNC, and Linear into "chramp".

For your last question - if a sample contains multiple instances of ecDNA and BFB, it can still be classified as ecDNA+, and this is what was done in both the Kim papers. However one caveat is that this creates an artificial heirarchy whereby biology driven by the BFB will be obscured by the ecDNA label. I recommend calling such cases "ecDNA+, BFB+" if you are able, however this can cause the resulting groupings to become more complicated. It is a decision you will need to make based on what makes the most sense for your cohort and the tests you are performing. :)

Thanks,
Jens

[GenomeArch]

Hi Jens,
Thank you for your detailed explanation. It has clarified a lot of points for me. I appreciate your insights on the classification of amplification features involved in distinguishing between ecDNA and BFB. I will consider your suggestion of labeling samples as "ecDNA+, BFB+" in the future.