# What percentage of people today would likely be descendants a man born 4,000 years ago? [closed]

Assuming a particular man was born in southern Mesopotamia, around 4,000 years ago, and was significant involved in a major religion.

1. To what degree (if any) of accuracy could his descendants be traced?

2. What percentage of the global population is likely a descendant of the man?

This question comes from a comment on sister site.

I don't think most people are really able to trace their entire family tree that far back. But Avraham is so many generations ago, I would expect that either nobody in the world is his descendant or nearly everybody in the world is

• I'm voting to close this question as off-topic because it is not about biology. – David Apr 28 '18 at 8:34
• If you are asking about any random person then it depends on a lot of factors and hence the question would be broad. If it is specifically about one person then it cannot be answered without having adequate and reliable archaeological evidences. Also, in that case the question is not really on-topic here. – WYSIWYG Apr 30 '18 at 13:00

To what degree (if any) of accuracy could his descendants be traced?

I don't really understand this question. To what degree can we trace its descendants given we know what from this man? Do we have a blood sample?

What percentage of the global population is likely a descendant of the man?

If $N_0$ was the population size 4,000 years ago and $N$ is the population today, then the expected number of descendent of any given person is simply $\frac{N}{N_0}$. However, there is a huge variance and this is where things get more complicated.

As coalescent processes are geometrically distributed (well approximated by an exponential distribution if we assume that the population is large; more info in this answer), then the variance in coalescent times is equal to the expected coalescent time.

The expected coalescent time in a diploid population of constant size $N$ is $2N$. Of course, the assumption of constant population size doesn't hold for humans due to the recent demographic explosion. When the population size is increasing the genealogy looks like a star as coalescent events are rarer and therefore, it is more likely for a given individual living in the past to live descendant today (Excoffier et al. 2009). However, I don't know how to calculate coalescent time in expanding populations.

• Remi, You could start by ruling out Indigenous Americans, Australians (and associated islands), and sub Saharan Africans. Then the Y-Chromosome is pretty conserved. Genetic anthropologists could likely look at the distribution of Y-chromosomes native to that region of the Middle East and then determine the frequency of that Y-chromosome appearing in samples through out the rest of the word. I don't think that you can really look at today's Raw population, as we have had too many depopulation events over the last several thousand years. Plague, wars, famine, monastic orders, etc. – AMR Sep 23 '15 at 5:34
• Working out the percentage of woman would be a total craps shoot. You don't have the option of using mitochondria, as the vast majority come from the mother, and with Meiotic recombination, working out the patrilineal X-chromosome would be near impossible. – AMR Sep 23 '15 at 5:37
• Does the $N_0 / N$ formula make sense? Should it be the inverse? What if $N_0$ is very small, don't you expect a given person from the 4 k years ago population to have more descendants? – Adrian Oct 22 '15 at 16:29
• Yes, you are right. Thank you for noticing the small mistake. I fixed it. – Remi.b Oct 22 '15 at 17:54