How do you determine genetic diveristy using 123 base pairs?

From Tooth gives up oldest human DNA,

If i’m reading this right, they’ve got six sequences of around 100 base pairs, and they’re comparing one sequence from 100,000 years ago to five others that are much more recent.

Yet somehow,

How can they say anything at all meaningful about genetic diversity of Neanderthals given so few, and such short, sequences of DNA, spred out over 70,000 years? Can they even tell if these sequences correspond to each other?

Bonus background questions for context:
About how many base pairs per typical human gene?
How many genes do humans have? (I have the numbers 100,000 and 30,000 arguing with each other in my head.)

I think this is the key. Mitochondrial DNA is very well conserved. The more those 6 sequences differ, the more it points to diversity. It doesn’t even have to code for anything.

Typically much larger then a hundred basepairs. Somewhere along the lines of a 1000-1500 would seem about right to me. But it doesn’t matter here since we’re talking about Mitochondrial DNA.


Probably somewhere in between.

I would think that using the mitochondrial DNA would say even less about genetic diversity. Mitochondrial DNA passes down through the mother only, so differences would only be an indication of how far apart those maternal lines were. Our chromosomal DNA is what gets intermixed in the gene pool and, to my understanding anyway, contributes to diversity that matters.

Also, I don’t think mitochondria DNA would affect very much how the organism develops. If you were to swap the mitochondria of, say, a poodle fertilized egg and a German Shepherd fertilzed egg, wouldn’t the poodle egg with GS mitochondria still develop into a toy poodle, and vice versa?

Yes, in all likelihood they’d develop the same. But that’s why you use mitochondrail DNA (m-DNA). The mitochondria function is so important, and so simple that there aren’t really many viable variations in the crucial parts of the m-DNA that affect how the mitrochondria functions. That means we can assume any variation is in non-crucial parts of the m-DNA, and is therefore basically random. That’s very important because it means we can assume that the more differences between two lineages, the longer ago they diverged.

If the DNA did affect how the organism works, then there’s a strong likelihood it will have some competitive (dis)advantage, which will result in faster changes than random (and we’d have no way of knowing how much faster, because we won’t know how strong the relative advantage is).

Not a good example of what you’re talking about because those are two animals are from the same species. Keep in mind that when you create a clone, you are only taking the nuclear DNA from one animal-- the mtDNA comes from the donor’s egg. same species mtDNA tansfer was problematic, cloning would be impossible (unless you used the donor’s own eggs).

I think we’ve done some cross-species cloning (ie, donor egg from another species), so it would be interesting to see how different mtDNA would have to be before problems would appear. My guess is that you could use mtDNA from fairly distantly related mammals and still have a successful clone.