The search for Denisovans in Southeast Asia islands: links with endemic super-archaic hominins and with modern humans

Remember the Denisovans: a sister lineage to Neandertals with a short fossil record, only composed of a few teeth and bone fragments from different levels of the Denisova Cave in Siberia – dated to between 50-250 thousand years (ka), and a mandible found in Karst Baishiya in the Tibetan Plateau – dated to c. 160 ka -, where Denisovan DNA has also been recovered from cave sediments in layers of 100 ka, 60 ka and possibly 45 ka.

From their DNA we know that Denisovans and Neandertals share a common ancestor, which in turn has a common ancestor with Homo sapiens. Denisovans also interbred with some Homo sapiens populations in East and South Asia, and this left traces in c. 5% of the genome of some present-day groups of Island Southeast Asia (ISEA), and notably of Papua New Guinea.

The question is: what causes the disparity between the lack of Denisovan fossils in ISEA, and the genetic evidence suggesting mixing events between modern humans and Denisovans in this region?

ISEA hosts a unique and rich fossil record of a variety of hominin groups, with a very open debate around the phylogenetic relationships between them:

  • Homo erectus, present in Java from c. 1.49 million years ago (Ma) until 117~108 ka.
  • H. floresiensis, an endemic species on Flores (Indonesia), which seems close to H. erectus, or alternatively to an even more archaic species that independently reached ISEA in a separate dispersal.
  • H. luzonensis, another endemic species on Luzon (Philippines). Its fossils share similarities with various hominin taxa including Australopithecus, Asian H. erectusH. floresiensis and H. sapiens.
  • Although we lack of Denisovan remains in the region, the multiple distinct pulses of Denisovan admixture in contemporary human populations of ISEA, New Guinea and Australia suggest that when modern humans arrived around 50 ka, they probably found at least one Denisovan-related group across ISEA, as well as local groups of H. floresiensis and H. luzonensis: a variety of hominin populations! 

Map of hominin occupation and megafauna survival in ISEA at the time of modern human arrival

Hominin occupation and megafauna survival in ISEA at the time of modern human arrival. Skulls=H. floresiensis & H. luzonensis. Stone tool=known artefacts associated with hominin presence. Buffalo and Komodo dragon=extant native megafauna. The estimated Denisovan ancestry in modern populations is shown in red in the pie charts, relative to that observed in Australo-Papuan genomes. Credit: Teixeira, J.C., Jacobs, G.S., Stringer, C. et al. (2021)


A new study explores the potential phylogenetic relationships between the various hominin groups when modern humans arrived the region:

1) Could H. luzonensis and/or H. floresiensis (or both) be the source of the Denisovan ancestry in modern human genomes in the region? The anatomical attributes of both species are not readily reconcilable with the few confirmed specimens of Denisovans from Altai and the Tibetan Plateau. Moreover, morphological and archaeological data suggest that the presence of both species preceded the estimated emergence time of the Denisovans. Same happens with Indonesian H. erectus, because of its last appearance date of 117–108ka.

2) Could H. floresiensis and H. luzonensis admix with the ancestors of modern populations now living in ISEA? Traces of interbreeding between super-archaic hominins and more derived hominin species have been detected in Altai Denisovans and, potentially, in modern Andamanese populations. The new study has searched 426 human genomes from across the world, including 214 individuals from Papuan and ISEA populations, for genomic signatures compatible with introgression from H. floresiensisH. luzonensis or other hypothetical late-surviving super-archaic hominin species. The results show no evidence of this mixing. However, modern human genomes carry traces of super-archaic ancestry that are embedded within introgressed Denisovan sequences (having previously been derived from ancient admixture events between Denisovans and an unknown super-archaic source), and these segments were detected in this study, as in previous works.


These islands are good candidates for future research efforts to recover evidence of the elusive ‘southern’ Denisovans…

  • The lack of any detectable super-archaic introgression in non-African modern human genomes, beyond trace levels indirectly inherited via past admixture with Neanderthals and/or Denisovans, stands in stark contrast to the strong evidence of Denisovan admixture with the ancestors of present-day ISEA populations.
  • Apparently, there was no interbreeding between H. luzonensis and H. floresiensis with modern humans, although interbreeding events could have occurred but the resulting lineages would have died out leaving no genetic traces.
  • H. luzonensis and H. floresiensis could have belonged to a clade that is closer to modern humans than previously thought, possibly being the late-surviving descendants of an earlier radiation of a Denisovan-like lineage across ISEA. The ubiquitous Denisovan ancestry in today’s humans could then be the result of modern humans’ admixture with one or both of these groups. The endemic island evolution (including dwarfism) of H. floresiensis and H. luzonensis would be complicating assessments of their morphology and possible phylogenetic relationships. However, this explanation does not align with the current consensus view that is based on interpretations of archaeological and fossil data found so far.


Reference: Teixeira, J.C., Jacobs, G.S., Stringer, C. et al (2021). Widespread Denisovan ancestry in Island Southeast Asia but no evidence of substantial super-archaic hominin admixture. Nature Ecology & Evolution.

Header image: Face of a Denisovan girl by artist Maayan Harel.

Acknowledgement: to Chris Stringer, for the information provided for this post.

More informationA new Denisovan! | Nutcracker Man

2 pensamientos en “The search for Denisovans in Southeast Asia islands: links with endemic super-archaic hominins and with modern humans

  1. Hi, very stimulating article. The map suggests an alternative scenario: 1) H. sapiens groups leaving Africa in an early “Out of Africa” wave of migration meet Denisovians somewhere in continental Asia (India?). 2) Later they reach and populate Australia as seafarer communities 3) Later again more H.s. peoples arrives from SE Asia in ISEA. The interbreeding with preexisting Australian people explains why the Denisovian ancestry is more present in the south. If this scenario is true, the Denisovian remains have to be searched in today’s Iran, Pakistan, India and Myanmar and not in ISEA.

    Me gusta

  2. There’s no such thing as other co-existing species of Homo. There simply hasn’t been enough geographic isolation time for new species to have emerged since Homo first appeared 2.6 million years ago. And humans in particular have an exceptionally low rate of mutation relative to other hominoid primates which makes it even more unlikely that new species of Homo could have emerged.

    So Homo has always been just one species. Homo habilis and Homo erectus should be viewed as chronospecies moving in the direction of Homo sapiens– no matter where they’ve existed.

    The higher level of genetic similarity between Eurasians, Neanderthals, and Denisovans is not surprising since they are descendants of the members of Homo that first radiated out of Africa probably around 2.1 million years ago.

    Researchers should really read Jolly’s insightful view on the subject:

    Am J Phys Anthropol

    . 2001;Suppl 33:177-204.
    doi: 10.1002/ajpa.10021.

    A proper study for mankind: Analogies from the Papionin monkeys and their implications for human evolution

    C J Jolly

    Me gusta


Introduce tus datos o haz clic en un icono para iniciar sesión:

Logo de

Estás comentando usando tu cuenta de Cerrar sesión /  Cambiar )

Google photo

Estás comentando usando tu cuenta de Google. Cerrar sesión /  Cambiar )

Imagen de Twitter

Estás comentando usando tu cuenta de Twitter. Cerrar sesión /  Cambiar )

Foto de Facebook

Estás comentando usando tu cuenta de Facebook. Cerrar sesión /  Cambiar )

Conectando a %s

Este sitio usa Akismet para reducir el spam. Aprende cómo se procesan los datos de tus comentarios .