Let me explain what I mean. The following is based on what I remember from high school biology, so please forgive the oversimplification.
We all have 46 chromosomes (23 pairs) in each of our cells that determine our genetic makeup. The exceptions are sperm and egg cells, which only have 23 chromosomes. Those 23 chromosomes are 1/2 of the 23 pairs of chromosomes, the 1/2 of each pair chosen (more or less) at random. A sperm and egg cell come together, creating a new entity with 46 chromosomes – the miracle of life!
We all know that in the case of identical twins, each twin receives the same 23 chromosomes from the mother and the same 23 chromosomes from the father, making them genetically identical. What if the opposite was true? What if there were two siblings, with the same mother and father, who received the exact opposite 23 chromosomes from their mother and the exact opposite 23 chromosomes from their father? I know the probability of this occurring is vanishingly small ( 1/2 to the 46th power, if I’m thinking about it correctly), but if that were to happen, could they even be considered genetically related?
Not completely genetically unrelated, no. Mitochondrial DNA is inherited only from the mother in humans and most other animals. If the siblings are both male, they inherit the Y chromosome from the father. There may be others that I am not thinking of as well.
You are correct that genetic relatedness varies by chance among siblings with the average being about 50% but it cannot be 0% based on sex linked DNA inheritance.
No, because both of them still inherit mitochondrial DNA from the mother only. You could do a DNA test on their mitochondrial DNA to prove that they had the same mother at least.
I read your post too quickly, and see your point about the mitochondrial DNA. So let’s say they can’t be 0% – what is the lowest percent “relatedness” that two siblings could possibly have?
Your understanding of probabilities of chromosome assortment is correct, but the entire process is even more complicated than that. Basically, when parents make their gametes, they cross over each chromosome pair to create new chromosomes. Usually there’s a handful of crossover events per chromosome pair, two or three on average IIRC. Since crossovers can happen just about anywhere on the chromosome (though not with uniform probability) the number of possible crossover products is astronomical.
So take your vanishingly small probability and raise it to an astronomical power. The resulting frequency will be so small that I feel comfortable saying that non-related siblings are practically impossible, even considering all sexually-reproducing creatures that will ever exist.
On further thought, you can estimate the probability of crossover results by picking three random spots out of the entire ~100 million base pair length of a chromosome. Google’s handy calculator tells me that 100,000,000 choose 3 ~= 10^23.
So we can multiply that by the probability of assortment to get (1/20^23) ^ 46 ~= 10^-1000 as a ballpark probability of unrelated siblings.
IIRC, I think it was Heinlein in one of his later novels (someone, anyway) describes a pair as “complementary siblings” where the genetic material was mutually exclusionary (not counting mitochrondrial, I asssume).
But, that was likely written in a more simplistic time; as lazybratsche points out, when the pairs of chromosomes duplicate, then can “mix and match” segments. Therefore, your children’s chromosomes are not duplicates (minus mutation) of grandparents’ chromosomes, but nice mix.
So the OP scenario of no shared DNA is possible, but extremely unlikely.
There’s a term: consanguinuity. There’s a quantification: Coefficient of relationship, but that’s normalized assuming typical share of genes between siblings.
But let’s take typical “DNA testing” as it would be used forensically. That would give you a probability that two samples are the same individual, or siblings, or whatever.
For all pairs of siblings, we could chart a histogram of the number they’d get in response to some standard forensic DNA test. There should be no entries at exactly zero %, and no entries at exactly 100%. But no doubt we’d see a more-or-less poisson-shaped curve, where most sibs fell in at a pretty high percentage, but some with fairly low readings.
So yes, it’s possible. How possible? That’s not really a sufficiently qualified question. You’d have to ask how possible is it to be below some specific percentage, and the answer to that is testable to a given degree of accuracy, if we can find enough test subjects.
It’s one of the tales in Time Enough for Love, and it didn’t happen randomly: It was the result of bioengineers deliberately separating out the chromosomes and sorting them that way. Which just might, barely, be doable with current technology, and is certainly no stretch of plausibility for technology centuries more advanced.
If we expand things a bit to “mostly genetically unrelated”, then the odds improve a bit. A friend of my dad had 3 daughters-two of them looked just like him (light colored hair, very roundish faces), while the 3rd looked just like his wife (thin face and dark hair). Many people (probably) thought the 3rd one was adopted.
I heard on Public Radio about a case in the U.S. where the mother was a chimera and had two separate sets of DNA so that one of her children appeared genetically unrelated to her.
The odds are surely long but the “opposite” sperm does get created as a part of normal spermatogenesis. The “opposite” egg is destroyed during oogenesis.
There are about 3 billion bases pairs in the human genome. There are hot spots, regions of the chromosome where recombination is relatively speaking likely roughly every 3000bp with very little crossing over elsewhere. So that makes for about 1 million hot spots where crossover is apt to occur. There is still a rather small, about 0.03% chance IIRC, of a cross over event at any given hot spot which makes for about 300 crossovers… quite a few, in the average human. This makes the odds very remote of ever duplicating the crossovers needed to make that “opposite” egg to pair with the “opposite” sperm which was already formed.
Lastly, hot spots are not discrete points but rather short regions, often of repetitive bases, which means that technically there are many millions of possible ways the genetic material could assort.
