If you read the latest post over at the Eurogenes blog, you'll notice in the comment thread a discussion by various researchers discussing ancient DNA. One of the researchers, "Chad Rohlfsen", is a researcher that is clearly associated the David Reich Lab at Harvard Medical School.
It's interesting to see that "Chad" thinks that when an ancient DNA sample has diverse or noisy components, he views it as a "jumbled mess": "Let's hope that the Jomon aren't a jumbled mess, like Kostenki."
It's possible that the sequenced DNA of both the Kostenki and MA-1 (Figure 1) samples were noisy, but it is just as likely that their component diversity represents a real, as yet to be explained, phenomenon.
Over and over again, these researchers continue to make the proposition that there was a time when populations did not experience admixture, or that admixture should be viewed as a "jumbled mess." They insist that tree-like splits in human populations are the norm, even when there is a lot of evidence that partial divergence and reconvergence of only partially diverged populations are also frequent events in human population history. (See for example, comments of Joe Pickrell ["ryukendo kendow"], insisting on tree-like splits in Europeans.)
Personally, I find it bizarre that Harvard is putting up with this.
List of some of the "researchers" commenting:
"Chad Rohlfsen" : a researcher associated with the David Reich Lab at Harvard Medical School
"Krefter" : Jean-Jacques Hublin with the Max Planck Institute,
"ryukendo kendow" : Joe Pickrell at Columbia,
"Davidski" : Wolfgang Haak at the University of Adelaide.