Mailbag: Neandertal-human comparisons

Re: Neandertal-human comparisons

Your website states, "of those positions where the human genome differs from chimpanzees, Neandertals have the chimpanzee version around 12.7 percent of the time."

Since the subject is the comparison with supposed MRCA of humans/chimps, shouldn't the correct statement be, "of those positions where the human genome differs from chimpanzees, Neandertals have the MRCA version around 12.7 percent of the time." ?

Or therefore, "of those positions where the human genome differs from chimpanzees, Neandertals have the chimpanzee version around 6.35 percent of the time."

If Neanderthals were something like 2 million base pairs closer to chimpanzee, shouldn't a few thousand of those base pairs be in at least a few modern Eurasians ?

Hi, thanks for your question!

Your point is correct that Neandertals do not have chimpanzee ancestors. If we were considering a comparison of all sites in the Neandertal sequence, you would be correct about the proportions. Neandertals would lack some proportion of the mutations that occurred on the modern human's lineage but they would lack every one of the mutations that happened on the chimpanzee lineage -- except for a very small fraction of parallelisms.

However, the comparison carried out by Green and colleagues was not of the entire genome, but specifically those sites in the genome that underwent mutations on the human lineage. The mutations on the chimpanzee lineage from the MRCA are completely ignored by this comparison.

The chimpanzee genome therefore stands in for the MRCA in this comparison. Sites at which both chimpanzees and humans have undergone parallel mutations have the potential to confound this comparison, because they are not counted (they are not places where the human and chimpanzee genomes differ). But the proportion of human substitutions that are also chimpanzee substitutions from the MRCA is very small, only around 1 percent of the human sites.

The fraction of Neandertal ancestry of Eurasians is around 3 percent, this is calculated differently, by examining polymorphisms within human populations today and considering the fraction shared by different humans' genomes with Neandertals. Eurasian people have around 3 percent more similarity with Neandertals than present-day Africans.