Melanesian genomes reveal episodes of interbreeding with Neanderthals and Denisovans

Study demonstrates multiple encounters with archaic humans

In a new attempt to obtain genetic information about Neanderthals and Denisovans, researchers have analysed the genomes of 1,523 genetically-diverse individuals, including 35 Melanesians. Results were compared with known Neanderthal and Denisovan sequences. 1340 Mb of the Neanderthal genome and 304 Mb of the Denisovan genome were obtained.

The Melanesians show between 1.9 and 3.4 percent of Denisovan ancestry. They have an average 104 Mb of archaic sequences: 48.9 Mb of Neanderthal, 42.9 Mb of Denisovan, and 12.2 Mb of ambiguous sequence that could be either. By contrast, only 0.026 Mb (in Esan) to 0.5 Mb (in Luhya) of archaic sequences per individual were found in Africans. An average 65.0 Mb of archaic sequences were found in East Asians; 55.2 Mb in South Asians; and 51.2 Mb in Europeans. Most of these archaic sequences were Neanderthal in origin, although a small fraction (less than 1 percent) in East Asians and South Asians are predicted to be Denisovan. There was evidence for an additional pulse of Neanderthal admixture in Europeans, East Asians, and South Asians compared to Melanesians. The data suggests that there were at least three separate episodes of interbreeding between Neanderthals and modern humans, and one of modern humans interbreeding with Denisovans.

The study also found a statistically-significant overlap between regions depleted of Neanderthal and regions depleted of Denisovan genetic sequences, suggesting that archaic sequences in these regions were deleterious and were purged by the effects of purifying selection. Regions depleted of archaic lineages are contain large numbers of genes associated with specific regions of the brain, particularly in the developing cortex and adult striatum. A large region depleted of archaic sequences spans 11 Mb on chromosome 7 and contains the FOXP2 gene associated with speech and language, as well as genes associated with autism.

It is likely that further studies will reveal an increasingly complex picture of how modern humans have interbred with archaic humans throughout Eurasia. The depletion of archaic sequences from brain-related sequences of the genome might hint at cognitive differences between modern and archaic humans; or these regions might simply be more prone to adverse effects of horizontal gene transfer.

Vernot, B. et al., Excavating Neandertal and Denisovan DNA from the genomes of Melanesian individuals (10.1126/science.aad9416) (2016).


Genetic study suggests that Sima hominins were proto-Neanderthals

430,000-year-old nuclear genome sequences confirms affinities  

Sima de los Huesos (‘Pit of Bones’)is a small muddy chamber lying at the bottom of a 13 m (43 ft.) chimney, lying deep within the Cueva Mayor system of caves in the Sierra de Atapuerca of northern Spain. Hominin remains were first reported there in the 1970s, and to date the remains of 28 individuals have been recovered. The Sima hominins lived around 430,000 years ago and while conventionally described as Homo heidelbergensis, they share some derived features with Neanderthals. This has led some to suggest that they are very early Neanderthals.

In 2014, mitochondrial DNA was obtained from the thighbone of one of the Sima hominins. It was expected that it would show affinities to later sequences obtained from Neanderthals, but instead it suggested that the Sima hominins were more closely related to Denisovans. However, mitochondrial DNA does not reveal the full picture of relationships among populations, so researchers set about the more difficult task of obtaining nuclear sequences from the Sima remains.

Genetic material was recovered from an incisor and a molar tooth, a fragment of a thighbone and a shoulder blade. Useful sequences were obtained from the incisor tooth and the thighbone fragment. The results have shown that the Sima hominins were, after all, more closely related to Neanderthals than they were to Denisovans. The Sima hominins were thus either early Neanderthals or closely related to the ancestors of Neanderthals after diverging from a common ancestor shared with the Denisovans. The age of the Sima remains is compatible with earlier estimates that the Neanderthal/Denisovan split occurred between 381,000 and 473,000 years ago. Based on the correctness of these estimates, modern humans diverged from Neanderthals 550,000 to 765,000 years ago – too early for later examples of Homo heidelbergensis such as Arago or Petralona to belong to a population ancestral to both Neanderthals and modern humans. The true common ancestor may be Homo antecessor, which was present in Spain from 1.2 million to 800,000 years ago and might have been responsible for the hominin footprints discovered at Happisburgh, England, in May 2013. However, this species has yet to be identified in Africa and may be a European variant of Homo erectus that migrated from Asia.

The Denisovan affinities of the mitochondrial DNA are still unexplained. One possibility is that the common ancestor carried mitochondrial lineages present in both, but later eliminated from the Neanderthals. The authors noted that this requires an explanation for the presence of two deeply divergent mtDNA lineages in the same archaic group, one that later recurred in Denisovans but disappeared from the Neanderthals; and one that became fixed in Neanderthals. The required explanation might be later population bottlenecks that are known to have affected Neanderthal populations. However, the authors preferred explanation is that the mitochondrial genomes of later European Neanderthals was acquired by interbreeding with hominins from Africa. This might explain the absence of Neanderthal-derived morphological traits in some European Middle Pleistocene hominins such as Ceprano and Mala Balanica.

Meyer, M. et al., Nuclear DNA sequences from the Middle Pleistocene Sima de los Huesos hominins. Nature (Published online) (2016).

Archaic human recovered from seabed off western coast of Taiwan

Could partial lower jawbone be from a Denisovan – or an entirely new species?

A partial fossil human jawbone from Taiwan is reportedly the first archaic hominin to be found there. The jawbone was dredged by a fishing net from the 60 to 120 m (200 to 400 ft.) deep Penghu Channel, 25 km (15.5 miles) of the island’s western coast. Also recovered were vertebrate fossils known as the terminal Middle/Late Pleistocene ‘Penghu fauna’. Both Taiwan and the Penghu Channel were part of the Asian mainland during Pleistocene episodes of lowered sea levels. The jawbone found its way to an antique shop in Tainan City, where it was purchased by a local man who in turn donated it to the National Museum of Natural Science of Taiwan.

The nature of its recovery means that there is no stratigraphic data by which the Penghu 1 jawbone can be dated. Accordingly, researchers measured its fluorine and sodium content in relation to that of other Penghu fossils. Fluorine, deriving from the surroundings, tends to accumulate slowly over time in buried bones; sodium on the other hand exists at about one percent in the bones of living vertebrates, but decreases when they are fossilised. By this means, the researchers matched Penghu 1 with fossil remains of Crocuta crocuta ultima, an extinct Eurasian subspecies of the spotted hyena that reached northern China between 500,000 and 250,000 years ago, but did not reach southern China until 240,000 years ago. There were episodes of lowered sea levels between 190,000 to 130,000 years ago and from 70,000 to 10,000 years ago; Penghu 1 probably dates to one of these two intervals.

Penghu 1 is identified as archaic by its relatively large molars and premolars, and by its lack of a chin. The short and relatively wide shape of its dental arcade is derived in comparison to the earliest humans (Homo habilis and the Dmanisi hominins), but other than that it cannot readily be assigned to any particular archaic human species. The second molar is larger than those of other archaic Asian hominins, and the low, thick body is closer to some examples African and European Homo from 400,000 years ago than to Early/Middle Pleistocene Asian Homo, with the exception of the 400,000-year-old Chinese Hexian Homo erectus remains.

The large second molar suggests Denisovan affinities in M2 crown size, but unfortunately no Denisovan lower jawbones or lower M2 teeth have yet been found for comparison. Not until we have a Denisovan lower jawbone that can be identified as such by genetic means will we have a better idea if Penghu 1 belonged to a Denisovan.  Nor can we rule out the possibility that Penghu 1 represents a completely new archaic human species.

Chang, C. et al., The first archaic Homo from Taiwan. Nature Communications 6, 6037 (2015).

530,000 years old Spanish hominins were closely related to Denisovans

Mystery of Sima de los Huesos ‘proto-Neanderthal’ mitochondrial genome.

Sima de los Huesos – ‘the Pit of Bones’ –  is a small muddy chamber lying at the bottom of a 13 m (43 ft.) chimney, lying deep within the Cueva Mayor system of caves in the Sierra de Atapuerca of northern Spain. Human remains dating to the Middle Pleistocene were first discovered there in 1976, and systematic excavation has been in progress since 1984. Investigation of the cramped site has proved to be long and difficult – it is located more than 500 m (⅓ mile) from the mouth of the Cueva Mayor and is hard to access, necessitating at times crawling on the stomach. To date, over 2,000 fragmentary hominin fossils have been recovered, including three skulls. In total, the remains are thought to represent at least 32 individuals of both sexes. Many of the remains are of adolescents and young adults, though, the pattern of mortality was probably quite normal for the time, and a similar peak in adolescence has been found at a site at Krapina in Croatia. There is no evidence for violence and the deaths could simply be the result of hunting accidents and childbirth complications. Hunting accidents were probably not uncommon among inexperienced young hunters and women likely fell pregnant soon after commencing menstruation (Pettitt, 2005).
Uranium-series dating suggests that the remains are least 530,000 years old (Bischoff, et al., 2007), and display a mixture of Homo heidelbergensis and Neanderthal features. For this reason, the  Sima de los Huesos hominins are often described as ‘proto-Neanderthal’ (Klein, 2009), although it has also been argued that they were a species distinct from both Neanderthals and Homo heidelbergensis rather than an intermediate between the two (Tattersall, 2002).

In a newly-published study, researchers at the Max Planck Institute for Evolutionary Anthropology have reported the sequencing of the almost-complete mitochondrial genome of one of the Sima de los Huesos hominins. The mitochondrial DNA was extracted from a thigh bone. An estimated age of 400,000 years was obtained by comparison with other, younger ancient DNA sequences dated by direct means. This is rather more recent than the uranium series dates for the site, but still by far the oldest hominin DNA ever recovered. The previous record-holder was no more than 100,000 years old.

Given the geographical location of the Sima de los Huesos and the apparent affinities of the hominins to Neanderthals, it was expected that the material would show affinity to genetic sequences obtained from later Neanderthal remains. Instead, it more closely resembled ancestral Denisovan mitochondrial DNA (Meyer, et al., 2013).

The Denisovan genome, first identified Denisova Cave in the Altai Mountains of southern Siberia, has been found in the modern populations of New Guinea and Island Southeast Asia, implying that the Denisovan range had once extended from the deciduous forests of Siberia to the tropics. This is a wider ecological and geographic region than any other hominin species, with the exception of modern humans (Reich, et al., 2011); but could their range have extended all the way to Europe?

It is likelier that the Sima de los Huesos hominins were the common ancestors of both the Neanderthals and the Denisovans. Mitochondrial lineages originally present in both lineages subsequently disappeared from the Neanderthals, but persisted in the Denisovans. They could have been lost from the Neanderthal line as a result of a population bottleneck of the type known to have affected later Neanderthal populations (Dalén, et al., 2012).


1. Pettitt, P., in The Human Past, edited by Scarre, C. (Thames & Hudson, London, 2005), pp. 124-173.

2. Bischoff, J. et al., High-resolution U-series dates from the Sima de los Huesos hominids yields 600 +/-66 kyrs: implications for the evolution of the early Neanderthal lineage. Journal of Archaeological Science 34, 763-770 (2007).

3. Klein, R., The Human Career, 3rd ed. (University of Chicago Press, Chicago, IL, 2009).

4. Tattersall, I., in The Speciation of Modern Homo sapiens, edited by Crow, T. (Oxford University Press, Oxford, 2002), pp. 49-59.

5. Meyer, M. et al., A mitochondrial genome sequence of a hominin from Sima de los Huesos. Nature (Published online) (2013).

6. Reich, D. et al., Denisova Admixture and the First Modern Human Dispersals into Southeast Asia and Oceania. American Journal of Human Genetics 89, 1-13 (2011).

7. Dalén, L. et al., Partial genetic turnover in neandertals: continuity in the east and population replacement in the west. Molecular Biology and Evolution 29 (8), 1893-1897 (2012).