Adam and Barbara have a son named Zeke. Later in life, Barbara has another child Yolanda. For this hypothetical, Yolanda’s dad is either Adam or Zeke (ick) and so if a test could eliminate one, we know the other is the father. The question is: with current genetic testing, how can we identify who is Yolanda’s father? The only thing I can thing of is that if Adam is the father, Yolanda will have exactly 1/2 of her genes from Adam but if Zeke is the father then she will have about 3/4 of her genes from Adam. Is something like that testable? Are there any markers that could be used to show one-generation vs two-generations away? Mitocondrial DNA seems like a no-go since Barbara is the mother of both Zeke and Yolanda.
An interesting question I’m not qualified to even take a WAG at.
You may have picked Yolanda to be a girl on purpose. I wonder how the question changes if Yolanda is instead Yaakov? Clearly we gain the ability to evaluate Yaakov’s Y-chromosome vs those of Adam and Zeke that Yolanda would obviously lack.
First, either there is a typo or your math is wrong. If Zeke is the father than Yolanda will have 1/4 of Adam’s genes and 3/4 of Barbara’s genes, and if my calculations are correct she will also have about 3/4 of Zeke’s genes
I don’t know the specifics of paternity testing, but provided there are enough genetic markers analyzed it should be not too hard to determine which was the father by comparing Yolanda’s DNA to any of the three relatives.
If we test vs Adam then we will call Adam the father if the genetic overlap is closer to 1/4 than to 1/2
If we test vs Barbara we will call Adam if the genetic overlap is closer to 3/4 than it is to 1/2
If we test vs Zeke we will call it Adam if the she has closer to 1/2 than to 3/4.
The main issue that I see is that it might require a custom analysis of the data. If you use an off the shelf kit, they might run their results through a fixed predictor kit that just spits out yes/no as to whether the genetic overlap is sufficient to indicate 1/2 or greater, and so wouldn’t work if tested against either Zeke or Barbara, since it would always show positive.
It depends on how thorough you’re willing to get with your tests. If, say, you were to completely sequence all of the parties involved (something that is well within our capability), then you could answer the question with near-absolute certainty.
Another test you could do, is that if Yolanda is the product of incest, then she’ll almost certainly have some duplicate chromosomes (or at least duplicate large sections of chromosomes, given crossovers).
Wouldn’t help, because if Zeke is Adam’s son, then they’ll both have the same Y chromosome.
But come to think of it, given that Yolanda is female, that does open up an option. If she’s Adam’s daughter, then she’ll have one X from Adam and one from Barbara. But if she’s Zeke’s, then her paternal X will be Zeke’s only X, which was Barbara’s. So checking for markers on the X, from Adam, Barbara, and Yolanda, will show whose she is.
I don’t know how much drift there is in a single chromosome in a single generation. But if we assume Zeke’s Y is an imperfect copy of Adam’s, and Yaakov’s is an imperfect copy of Adam’s or Zeke’s as the case may be, we might be able to get some metric of drift if we looked close enough.
If nothing else, finding a matching change in both Zeke & Yaakov would suggest it occurred on the Adam → Zeke transition and was faithfully copied to Yaakov rather than independently arising in the same way in the same spot in Adam → Zeke and Adam → Yaakov.
As you say, it is practical to sequence all four genomes completely if we cared to go to that much trouble and expense. So these comparisons aren’t ruled out in principal.
Its probably going to be pretty minsucle. If you did a whole genome sequencing you might be able to pick up something, but if you using a standard paternity test with a limited number of targets then no chance.
I did not pick Yolanda to be female specifically. My original thought was Yancy but that is an interesting question if there is a difference in the analysis for male offspring vs female offspring.
Zeke gets 23 chromosomes from Adam and 23 from Barb (the complementary pairs except X and Y.)
Yolanda would get the same if Adam’s child, except she does not get the exact same 23 from Barb - let’s say 11 or 12 Barb-genes would match with Zeke’s barb-genes. One of her X’s may or may not match Zeke’s.
However, overlap with Zeke is 50-50, so the other 23 Adam-genes may or may not his match also - say aboout 11 match Zeke.
So Yolanda would match half Barb’s genes and half Adam’s genes with Zack’s genes.
If Zeke’s child - 23 from Barb. One would be an X that may or may not match Zeke. The 23 Yolanda gets from Barb, like case above, maybe 11 or 12 match Zeke’s genes from Barb.
From Zeke, she would get 23 genes. Approximately (odds say) 11 or 12 will match Barb’s genes, and 11 or 12 match Adam’s genes.
So Zeke’s child will have about 33-34 of Barb’s chromosomes (3/4) and 11-12 of Adam’s (1/4). But she will have 50% (23) of Zeke’s.
So Adam would test as equivalent of a grandparent (duh), half sibling, uncle/aunt, etc. If Zeke were the father. Presumably Barb’s test would point to inbreeding anomalies with Yolanda.
Then there’s the whole issue of genetic crossover, where chunks of one of those 23 pairs can switch off with each other during cell division - thus confusing the math. I’ve heard claims this happens frequently so a few generations of inheritance, the chromosomes are all shaken and stirred part A and part B.
Since the legal paternity tests specifically are intended to point to 50% match, for forensic reasons, I would assume that they are designed to rule out uncles, grandparents, etc. Not sure how extensive the 23-and-me relative/family tree stuff is supposed to be, since presumably the purpose is to show whether someone is even a distant genetic match, not replace family frolic detection. It will tell you whether the gardener or milkman did it, but will it tell you if the culprit is a close relative?
You can identify this kind of ahem pedigree collapse via autosomal DNA testing, so yeah, 23&me and similar services will give you the data that can be correctly interpreted to identify incest within a couple of generations.
You can even download the raw DNA from 23&me et al and upload to Gedmatch, where they have a specific tool to identify incestuous descent. You would only need Yolanda’s DNA; not any of the parents.