How Many Differing Snv To Expect Between Two Cancer Samples With Identical Origin?

Question

[fubioinf]

Can anybody tell from their own experience how many different SNVs in general and missense novel SNV calls to expect between two cancer cell line samples that once were derived from the same "mother" cancer cell colony or in general when comparing cancer cells? I.e. as well SNVs that you would expect both cells to share but in fact they don't anymore, as well as new SNVs that either cell might have acquired over a period of time, distinguishing them from another.

Personally I've identified 110k up to 330k many different SNVs in the genome between distinct pairs of samples of cancer cells with identical origin depending on their treatment and I would like to interpret that number more confidently.

Answer 1

[Chris Miller]

The number of passages each portion has gone through will have a huge impact on how many mutations you're likely to see. Off the top of my head, over 100k sounds high, but if these are widely diverged from the ancestral line or have repair defects, all bets are off.

Answer 2

[Giovanni M Dall'Olio]

You can have a look at the paper of the HeLa genome, published a few months ago. The HeLa cell line is the most famous cell line available to date, and during the last 30 years has accumulated a lot of mutations, so I would consider it as an upper limit for the number of SNVs to expect.

In the paper, they found 1,750,535 SNVs homozygous SNVs and 3,026,053 heterozygous SNVs - I am not sure how this relates to your question, though.

[Chris Miller]

To be clear, that's 3 million sites that differ from the reference genome, which is pretty typical for any human genome. If you're looking at cells with the same recent progenitor, this number seems quite high, unless they've been grown independently for a long time.