A few years ago a paper was published, Effect of ancient population structure on the degree of polymorphism shared between modern human populations and ancient hominins, in PNAS, which argued that ancient population structure within Africa might be the reason that non-African populations are genetically closer to Neandertals. The basic logic is simple. If within Africa there was variation in relatedness to Neandertals, and modern non-Africans are derived from the group that was closer to Neandertals, then one might infer that there was recent admixture between the groups even though any connection was very distant in the past. This objection actually popped up immediately when the Neandertal admixture Science paper was published. Over time many people have been convinced by various ingenuous and abstruse arguments.
The problem is that not everyone is a statistical geneticist, nor do they think about these issues very often. When it comes to the media they have to rely on what’s being published in prominent journals. The PNAS paper above has haunted the field in my opinion, because reporters have had to take the researchers at their word, when frankly most statistical geneticists that I know did not find their arguments very persuasive in the first place. Today we have even less reason to believe them. A few hours ago I saw this in my feed from the Genetic Literacy Project, More mystery about Neanderthal and modern humans: How reliable is ancient DNA analysis? The answer is very reliable. There’s no controversy.
The figure at the top of this post is from Genome sequence of a 45,000-year-old modern human from western Siberia. What you see is that the markers diagnostic of Neandertal ancestry are clustered together in segments to a far greater extent in the ancient sample than among modern peoples. This sort of pattern of decaying tract length is a hallmark of pulse admixtures. Each generation recombination breaks apart associations until the tract lengths are very small (or, they are not detectable). There are now at least two ancient samples which show that Neandertal-modern human admixture occurred in the relatively recent past in relation to their own period in comparison to modern individuals. That in itself reduces the probability of ancient structure being the dominant explanation for the Neandertal affinities of non-Africans, as ancient structure would not exhibit this tract length bias. And, this result has the utility of being amenable to common sense comprehension.
Comments are closed.