While sampling some nice beer in Århus earlier this year with my good old friend Thomas Mailund, we had an interesting discussion about how long our genes live on for.
I was reminded of this discussion when I managed to find my great-great-great-great-great-great-great-great-great-great-great-grandfather on Google. His name was Georg Widmann, and he was born around 1532 in Heiningen in Württemberg.
However, as I discussed a few years ago, we get half our genes from our father and the other half from our mother; we therefore get 25% from each grandparent, 12.5% from each great-grandparent, etc., so when do we reach a point when there’s nothing left?
According to Wikipedia, the “haploid human genome contains ca. 23,000 protein-coding genes”. 23,000 can be halved 14.49 times, which equals around 435 years (at a generation length of 30), and this means that any ancestor born before the year 1575 is likely to have contributed less than one gene to our genome. (Georg is of course an exception – ignoring the possibility of adultery and mutations, my Y chromosome is an exact copy of his.)
One can look at the numbers differently, too. Genes are defined by the “2.9 billion base pairs of the haploid human genome”. 2,900,000,000 can be halved 31.43 times, taking us back 943 years to the year 1157, but that will include ancestors who have only contributed junk DNA.
The “human genome contains vast regions of DNA the function of which, if any, remains unknown. These regions in fact comprise the vast majority, by some estimates 97%, of the human genome size.” 3% of 2.9 billion base pairs is 87 million base pairs, which would take us back 791 years to the year 1219.
However, the “nucleotide diversity between humans is about 0.1%, which is 1 difference per 1,000 base pairs.” That would take us down to 87,000 base pairs that actually matter, and that number can be halved 16.41 times, which would take us back 492 years to the year 1608.
To conclude, I’m not absolutely sure what the cut-off point should be. There’s definitely no point in doing genealogy further back than the year 1157 (except for pure patrilineal and matrilineal descent), but there are good arguments also for stopping in 1219, 1575 or 1608.