Understanding Genetic Similarity: Dawkin's Perspective on Family Lines

[Phobos]

In Dawkin's book "The Selfish Gene", he discusses genetic relationships in family lines (e.g., your genetic similarity is 1/2 that of one of your parents, 1/4 of your grandparents, etc.). He mentioned that going back only a few generations, your genetic similarity to your family line is no different to than it is to the rest of the population.* At what point is that genetic similarity reached? In other words, what is my genetic similarity to the average human genome? In other other words, how many great-great-greats do I need to trace back before my ancestor is no different than a total stranger?

* Making extensive genealogy somewhat pointless?

 

[matthyaouw]

Thats a very good question. It could finally help us answer the age old problem- how distant must a relation be before I can sleep with her without it being immoral?

 

[ryokan]

As in chess?

In Dawkin's book "The Selfish Gene", he discusses genetic relationships in family lines (e.g., your genetic similarity is 1/2 that of one of your parents, 1/4 of your grandparents, etc.). He mentioned that going back only a few generations, your genetic similarity to your family line is no different to than it is to the rest of the population.* At what point is that genetic similarity reached? In other words, what is my genetic similarity to the average human genome? In other other words, how many great-great-greats do I need to trace back before my ancestor is no different than a total stranger?

* Making extensive genealogy somewhat pointless?

That remember the classic prize to the chess inventor (although with negative exponent). In this case there would be a dilutional effect of similarities growing exponentially with the number of generations going back. I don't know which is the % of allelic variation among human beings. Such knowledge could allow to calculate n. Are you in agreement?

 

[Phobos]

I’ll take a stab at answering my own question…

Seems that different DNA sequences have different frequencies in the human gene pool, so perhaps a general comparison for the whole genome may be a bit tricky. But I seem to recall hearing from the Human Genome project that the similarity in genes of nuclear DNA between any two people is about 99.9%. That’s a 1/1000th difference, which works out to about 10 generations (1/2 gene similarity to parents, to grandparents, 1/8 to great grandparents, etc.)

Sound about right? Or is that 99.9% comparison too simplistic?

10 generations is only about 200-400 years of history...so perhaps genealogy is not a lost cause after all. :)

matthyaouw - I'll leave the morality question to you. :) But my calculation assumes no imbreeding...which is probably a poor assumption given the historical tendency for people to marry their cousins.

ryokan - Sure. My calculation above for determining 'n' generations is 1/(2^n) = 1/1000...so, n is about 10.