Absolute and relative Copy numbers, tumor and ploidy

Question

[ashenflower]

Hello everybody,

I started working with cancer data, in particular with CNAs. Unfortunately, I do not have a strong biological background, and basically none in cancer. So, even after reading all the related answers I could find 1 , 2 , 3 , 4, and more related papers, I still have many doubts. I will try to list them for the sake of a clear exposition:

1. How is the CN defined? I'll try to explain. When detecting CNVs using read depth, what is usually done is mapping back the reads and then try to identify regions with higher/lower coverage then others. So, for e.g., in the case of a normal diploid genome, if I had a copy of a region only in one chromosome, by mapping back all my reads to the reference, I would expect to find that one region has double the normalized coverage of the others, so detecting that there is 1 more copy of that particular region with respect to a reference. In the case of a tetraploid sample, I would expect that if a region is duplicated in a single cromosome, I would still detect an absolute gain of 1 for that region. Is this correct?
2. What are the relative and absolute copy numbers? I found these two answers on the web 5 6, but they look differet to me, so I couldn't figure out. Taking into account the previous example, i.e. having a gain of 1 copy of a segment, would this be a relative CN? And if I knew the ploidy, how would I use it to find the absolute CN?
3. In my case, I would need to compare the results from multiple tools. Would it still make sense to eventually compare the relative CN, if I do not know the tumor ploidy?

Thank you in advance for any answer.