To my understanding, one needs the B-allele frequencies for allele-specific copy number. To get those, you usually use germline SNPs for the following reason:
"In this context, the b allele is the non-reference allele observed in a germline heterozygous SNP, i.e. in the normal/control sample. Since the tumor cells' DNA originally derived from normal cells' DNA, most of these SNPs will also be present in the tumor sample. But due to allele-specific copy number alterations, loss of heterozygosity or allelic imbalance, the allelic frequency of these SNPs may be different in the tumor, and that's evidence that one (or both) of the germline copies was gained or lost during tumor evolution."
However, is there anything wrong with using somatic SNVs as well? Somatic SNVs are confirmed heterozygous sites that should give the same insight into frequencies of the non-reference allele, right?
There will be a lot less of them, so the impact will be negligible.
But wouldn't it be better to use both somatic SNVs and germline SNPs even still?
With germline, you know the original frequency is either 50% or 100%. With somatic, you do not. As you said, somatic variants are still informative. They can be used to estimate tumor purity and clonality, and there are specialized tools for that.
When you say "With somatic, you do not", so it is because of tumor purity and clonality that deviate this?
Yes. You also don't know the absolute copy number. For example, if the variant is at 25%, it could be 25% tumor (homozygous), 50% tumor (heterozygous), amplification of the non-mutated copy (1 mutant, 3 WT), some combination of those, or a more complex event.
What about germline SNVs (not SNPs)?
Germline SNVs are SNPs
My mistake. I thought the distinction was SNPs are found within a certain percentage of the population, whereas germline SNVs could still be below that percentage.
For the purpose of somatic/germline discussion, any germline/inherited SNV is a SNP and tumor-specific is somatic. For population genetics, you can have common and rare SNPs. They are germline either way.