Preview Mode Links will not work in preview mode

Razib Khan's Unsupervised Learning

Jul 14, 2021

Last week two new hominin fossils were published in the scientific literature, and extensively reported on in the media. “Dragon Man”, discovered in Harbin, China, and dating to 140,000 years ago is claimed to be a new species that is the closest to the modern human lineage. Meanwhile, the hominin discovered at Nesha Ramla in Israel dates to 120,000-140,000 years ago, and it seems most similar to Neanderthals (though its tools are no different from modern humans to the south and west in Africa).

I’ve given some thought to the implications of these results, and how to interpret them. But I wanted to get the sense of another geneticist, my friend Vagheesh Narasimhan. I’ve talked to Vagheesh before in relation to his blockbuster paper, The formation of human populations in South and Central Asia. My goal for this podcast was to “nerd out” on human evolution from a genomics perspective, and see if he had the same impressions that I did of these papers.

Vagheesh Narasimhan

We discussed population structure in Denisovans, the importance of ancient DNA and proteins, whether the “Out of Africa” theory even makes sense, as well as new work on methylation patterns in the genome and predicting physical characteristics. I did bring up statistical power in “skull science,” and both of us expounded on why DNA, in particular, is so powerful as a method of inference in comparison to traditional morphology.

It was an hour spent slashing back and forth across these two papers and circling around from a genetic perspective.

Both of us agreed that we can’t conclude Dragon Man is closest to modern humans.

Past paleoanthropology podcasts: