mitochondrial variants in familial disease

Question

[peris]

Generally un-related patients are used to study role of mitochondrial SNP on human disease. Now, if we need to study role of mitochondria SNP on familial disease (affected siblings or cousins) what logic we need to use because there will be high likely that majority of the mitochondrial SNP will be same between siblings.

Answer 1

[Ying W]

You would need to model the mito inheritance (mom -> child) between family members.

[peris]

You want to say I should filter out the variants which are present in both mom and child?

[Ying W]

Its a bit more complicated than that because instead of 2 chromosomes that you get from mom+dad, you will have thousands of mitochondria that you get from egg (mom) along with possibly some from sperm. The mitochondria will probably be a heterogeneous population with some proportion having mutations that others do not have. Some medical conditions only become penetrant when the proportion of mitochondria with mutation is high enough. Sequencing is taking a population of cells from individuals and sequencing it, the mutations in mitochondria that you are assaying is a mixture of thousands of mitochondria per cell from millions of cells.

So you would want some estimate of how often a mitochondrial mutation shows up (to do it right, you would need a mutation caller with non-diploid assumptions) and see if prevalence of mutation is associated with disease while keeping in mind that the prevalence should be similar for members with shared maternal ancestry.

[peris]

Thanks for your explanatory reply. I am using mtoolbox for calling the mt variants.