How to quantify the correlation of variants between cells?

0

Entering edit mode

3 months ago

zau saa ▴ 120

Hi!

I'm going to compute the distance between cells by their variants, but I have no clue.

For example, here are variants of mitochondria from two cells:

mttype_1    A73G,A210G,230delA
mttype_2    A73G,220delT

What's the distance between these two mttype?

variant • 471 views

ADD COMMENT • link 3 months ago by zau saa ▴ 120

0

Entering edit mode

Can you please explain the details of the problem behind this task and more information about the size of real data, how many samples and how many variants, how have the variants been detected and why are they in this format and not in VCF for example? There is no way to establish the correlation (or linkage) of these variants with only two samples, but then what do you really want distance or correlation? If you are doing some sort of phylogenetic approach, a distance matrix is not doing much good either. So what is the problem you are trying to solve? (obviously, the edit distance between the two sequences is 3 if there were no other differences)

ADD REPLY • link 3 months ago by Michael 54k

0

Entering edit mode

Thanks for your reply! I identified about 30 mitochondrial DNA types from the scDNA-seq data of a sample before and I wonder if these types of mitochondrial DNA are related. (e.g. whether they have the same mutation direction)

ADD REPLY • link 3 months ago by zau saa ▴ 120

1

Entering edit mode

Why not do a standard phylogenetic approach first? Add an outgroup and see how they form clades (or not).