5
$\begingroup$

Descendant of n generation has on average 1/2n DNA of ancestor. (For example children have 1/2 DNA of parents and 1/4 DNA of Grandparents, After 10 Generation 1/1024 DNA and after 100 Generations 1/2100 DNA).

Is there a point (value of n) when descendant is no more related to ancestor and is similarly related to random individual from an ancestral population.

$\endgroup$
2
  • 2
    $\begingroup$ Welcome to Biology.SE. Your question sounds like a Coalescent theory question. Do you assume that no mutations and no selection in your model? The answer will very likely depend on the population size. $\endgroup$
    – Remi.b
    Jan 26, 2015 at 7:33
  • $\begingroup$ Thank you. Answer will likely depend on effective population size. logically mutations will decrease value of n and selection will increase or decrease depending how it favors certain DNA. I just did not thought about this and wanted to know if this is even possible, that descendant is no more related to ancestor than to random individual from an ancestral population. $\endgroup$ Jan 26, 2015 at 8:13

1 Answer 1

2
$\begingroup$

There is an interesting way to look at this question. By definition, two individuals are related only if they have a common ancestor. (If you mean something else by it, please correct me. For completeness, I include the trivial case in this: that is, I regard you as one of your own ancestors.) So your question can be inverted: it is really asking after how many generations everyone in the ancestral population is your ancestor?

As indicated in the comments, the answer will depend on the specifics of the population structure, including its size. But a first stab can be taken in this way: you had 2 ancestors one generation back, 4 two generations back, ..., $2^n$ (not necessarily distinct) $n$ generations back. So you certainly cannot be descended from everyone in the ancestral population if less than $\log_2 N$ generations have passed. ($N$ = population size, which for simplicity I assume to be stable.) In a freely mixing population under the very simplest assumptions, you will eventually decay exponentially towards complete coverage with a half-time of one generation.

This is complicated by the fact that some lineages die out completely. For instance, some of the individuals in the ancestral population didn't have any children. Some had children, but no grandchildren, etc. Obviously none of them were your ancestors, so you will never be related to them (except through previous common ancestors), no matter how much time passes. In fact, as time passes, more lineages tend to die out. So, if you look back very far, you'll find that the entire current population has only a very small numbers of ancestors in the distant past. This is called coalescence, and is what people are talking about when they speak of "mitochondrial Eve", for instance.

So, in summary, it's complicated. It takes only a fairly small number of generations ($O(\log_2 N)$) before you're potentially related to most of the ancestral population. But over longer times ($O(N)$ generations) your relationship to the ancestral population actually goes down.

$\endgroup$
2
  • $\begingroup$ Thank you for answer, very interesting. My question is slightly different, will make it more clear: Is my DNA more similar to Mitochondrial Eve or Y Adam or it is not possible to distinguish from other individuals who live with Mitochondrial Eve or Y Adam. I understand that Mitochondrial DNA and Y chromosome is not recombining, but they have higher mutation rates than autosomal DNA and also, they are just a small part from total DNA. $\endgroup$ Jan 28, 2015 at 12:24
  • 1
    $\begingroup$ Your mitochondrial DNA is more similar to that of mitoEve than to her contemporaries. Your Y chromosome is more similar to Y Adam than to his contemporaries. Other regions of the genome will trace back to one of them, or to other humans living in the distant past. It's hard to say a lot more without making detailed assumptions about the genetic structure of the population. $\endgroup$
    – Leon Avery
    Jan 28, 2015 at 14:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .