View Full Version : Does the number of ancestors one has directly affect one's DNA?

05-03-2019, 11:49 AM
Please suppose, for argument's sake, that human A and human B have the same genealogical history in every way, except for the fact that human A has twice the number of ancestors as human B.

All of A's ancestors (as well as A himself) were born to parents who were x years old, whereas all of B's ancestors (and B himself) were born to parents who were 2x years old.

Will there be any differences between human A and human B that are solely related to the number of ancestors and not anything else?

(So please disregard anything that may be attributed to differences in the ancestors' ages, e.g. anything handed down specifically due to the age of the mothers. I would just like to know if the sheer number of ancestors, in and of itself, has any bearing on the DNA of an individual.)

05-03-2019, 07:52 PM
By the same genealogical history do you mean.
1. They both have a great great grandfather who was Chinese and the rest of the great great grandparents were English
2. They both have a Chinese ancestor from 1850 and the rest of the ancestors from 1850 were English.

05-03-2019, 08:10 PM
Number one.

05-05-2019, 07:25 AM
Unless I'm misunderstanding your query, yes, but you can't discount the age of the ancestors, as that's implicit in the difference.

Person A and B will have slightly genome profiles because of germline mutations. The older you are, the more mutations (in general) you're likely to accrue. That includes mutations which specifically affect your offspring (germline). In your example, Person B is statistically more likely to have additional number of mutations in comparison to Person A, given both of their parents were older. Worth pointing out that the germline mutation rate in males is about four times greater than in females (https://www.nature.com/news/fathers-bequeath-more-mutations-as-they-age-1.11247)(just my informed opinion, but that's probably evolutionary mechanism, parallel to the limited number of ova females carry, which assures generational propagation i.e. "out with the old, in with the new", "don't overstay your genetic presence" etc.).

There's also an increased risk of chromosomal translocations the older an individual is (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3025794/). About 1 out of every 500 live births carry a reciprocal translocation (where two adjacent chromosomal arms intersect and swap DNA). These aren't generally thought to produce an adverse affect on the child's phenotype.

In the context of genetic genealogy, this will definitely lead to a difference in Y-chromosome testing (per above). Aging should also theoretically increase mtDNA mutation acquisition (definitely so for somatic cells (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1762815/), I would infer so for the germline). I suppose it could result in differences in ancestry approximation if enough generations (dozens?) with older parents take place to form Person B.

In short, assuming all other factors are equal, the increased generational span for Person B will confer additional mutations across the board relative to Person A, and the odds of a measurable difference across all standard genealogical parameters will increase with each successive generation.

05-06-2019, 04:48 PM
Thank you, DMXX. (I would've replied sooner, but I was not notified by email even though I have it checked off.) So then the total number of ancestors means nothing, and it is only the ages that matter. I wondered if A would somehow be more evolved, or just somehow genetically different, since he was the result of twice as many reproductions as B. I thought more reproductions would mean more chances for mutations in general.

05-06-2019, 05:37 PM
Okay. I assumed the question was geared towards genealogy (hence half of it focusing on Y-DNA and mtDNA) rather than an abstract, mutation-driven, Darwinian view of reproduction. This requires a reframing of my answer:

The total sum of mutations is different from evolutionarily useful mutations. As you've stipulated that all other conditions or factors between the two people are the same, there's no mechanism to cause one individual to inherit an evolutionarily useful set of mutations with the passage of time in comparison to the other. By virtue of your stipulation (which is somewhat independent of the generation count, unlike parent age), neither of your hypothetical people should become "evolved" in a different direction from the other, given their environmental factors are the same.

Saying that, merely doubling the reproduction rate down one line would probably increase the frequency of translocation events, but in vivo mutations during fertilisation (de novo mutations)? Those definitely happen, where about 70 SNV replacements happen per generation (https://www.nature.com/articles/nrg3241), among other changes (note: SNV's differ from SNP's (https://www.sciencedirect.com/science/article/pii/B9780124047488000083)). That'll be the main driver of change for Person A relative to B.

So, in this hypothetical, you're dealing with one individual who's exceptionally prone to de novo mutations, and another who's exceptionally prone to germline mutations. The answer remains "yes".

If there aren't any selective pressures in this vacuum, then they won't develop along any particular Darwinistic trajectory. In reality, there likely are selective pressures applying to both individuals, and they probably wouldn't be the same throughout a given period of time (say two millennia). So, in a more realistic example, Person A would have "evolved" along a trajectory in response to the various selection pressures faster than Person B would have.

Who's going to accrue more mutations, full stop? That sounds like a fun weekend maths exercise (remember that both Person A and Person B are liable to de novo mutations).
If I were to guesstimate with some evening, been-up-for-13-hours cognition, Person B is probably the beneficiary of a larger sum of mutations (assume Person A and B's ancestors were all 20 and 40 years respectively, assume a one millennia timeframe, assume negligible germline mutations for Person A, assume an average of 45 germline mutations for Person B; 2^50/2 * ~70 vs. 2^33/2 * (~70 + ~45); latter feels larger).

05-07-2019, 11:58 AM
Excellent. That's exactly the type of explanation I was hoping for. Thank you DMXX.