A recent research article (not peer-reviewed) published in BioRxiv (Scalable ARG-free Detection of Denisovan-mediated Superarchaic Introgression Reveals Heterogeneous Patterns across Populations. Noel McAllister, Sebastian Zöllner, Xinjun Zhang, bioRxiv 2026.06.25.734355; doi: https://doi.org/10.64898/2026.06.25.734355), used simulations based on a specific computing program to explore the sites in our genomes where superarchaic introgression may be found. Introgression that reached us via Denisovans.
It looks into the genome of modern populations and finds that HLA/MHC loci have higher introgression scores than other sites. This, according to the authors, "...is consistent with previous evidence that immune-related loci retain archaic variants. At the same time, HLA/MHC is among the most polymorphic regions of the human genome and has been shaped by pathogen-mediated and balancing selection, both of which can preserve deeply divergent haplotypes and generate unusually deep local genealogies..." However, they caution that these high scores may not only be due to archaic introgression: " ...HLA/MHC is also one of the clearest examples of long-term balancing selection, which also preserves deeply diverged haplotypes without recent introgression... We therefore interpret the HLA/MHC signal cautiously. It should not be treated as definitive evidence for superarchaic ancestry by itself. Nonetheless, its repeated elevation across all three populations is notable and suggests that immune-related regions may be especially informative targets for future local haplotype, ARG-based, and functional analyses."
It makes sense that alleles linked to immunity, received from superarchaics, if they provide a benefit, will be preserved across generations.
The paper then discusses Denisovans and mentions an interesting fact, that Denisovans could have evolved from Homo erectus: "Our results also intersect with the recent paleoproteomic evidence suggesting that some Denisovan superarchaic ancestry may ultimately derive from populations related to H. erectus."
It also notes that population structure can be mistaken for introgression: "A further limitation is that deep population structure and introgression remain intrinsically difficult to separate. Structured ancestral populations can generate old coalescent times and divergent haplotypes, and some recent models have shown that deep structure may explain signals previously interpreted as archaic or ghost admixture."
It presents a scenario that I had not read about before: "Hominin evolution may have involved both long-lived structured populations and episodic admixture among deeply diverged groups." Which is very plausible, we can imagine, especially in Africa this scenario, with archaic populations and ancestral modern humans living in isolated, deeply structured populations admixing sporadically, and adding diversity to the Homo sapiens group.
The other branches of the human tree also followed that pattern, with Neanderthals in western Eurasia and Denisovans in the Eastern part of Asia. Mixing with modern humans as our ancestors met them.
Homo erectus has been shown to have admixed with Denisovans (see this post), but, this article suggests that Denisovans evolved from Homo erectus, which is a very reasonable assumption. I recently posted about Denisovans and Homo heidelbergensis being related (also Neanderthals), it is possible that the common ancestor for these groups evolved from erectus (via Homo antecessor) in Eurasia.
Evidence seems to be building a stronger case for an Eurasian origin for hominins.
Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2026 by Austin Whittall ©
