Wednesday, April 18, 2018

Progress Made On 68 Year Old Math Problem

How many colours are needed to colour the plane so that no two points at distance exactly 1 from each other are the same colour? 
This quantity, termed the chromatic number of the plane or CNP, was first discussed (though not in print) by Nelson in 1950 (see [Soi]). Since that year it has been known that at least four and at most seven colours are needed. 
The lower bound was also noted by Nelson (see [Soi]) and arises because there exist 4-chromatic finite graphs that can be drawn in the plane with each edge being a straight line of unit length, the smallest of which is the 7-vertex Moser spindle [MM] (see Figure 7, left panel). 
The upper bound arises because, as first observed by Isbell also in 1950 (see [Soi]), congruent regular hexagons tiling the plane can be assigned seven colours in a pattern that separates all same-coloured pairs of tiles by more than their diameter. 
The question of the chromatic number of the plane is termed the Hadwiger-Nelson problem, because of the contributions of Nelson just mentioned and because the 7-colouring of the hexagonal tiling was first discussed (though in another context) by Hadwiger in 1945 [Had]. The rich history of this problem and related ones is wonderfully documented in [Soi]. Since 1950, no improvement has been made to either bound. 
From here.

The author, a professional scientist and amateur mathematician, proves in the linked pre-print that the lower bound is 5 and not 4.

So, the correct answer is now 5, 6 or 7, although we still don't know which of those three integers is the correct answer.

Saturday, April 14, 2018

Neanderthals And Modern Humans Experienced Hybrid Incompatability

Also know as Haldane's law, heterozygotes for sex determination genes are strongly disfavored as children of cross-species hybrids. And, the evidence is strong that this hybrid incompatibility prevented Neanderthal-human hybrids from being born. This is strong evidence that Neanderthals and modern humans were biologically separate species.

As the author explains in an open access summary:
After years of sequencing the genomes of female Neandertals, researchers have finally got their first good look at the Y chromosome of a male Neandertal—and found that it is unlike that of any other Y in modern humans living today. Even though Neandertals and modern humans interbred several times in the past 100,000 years, the DNA on the Y chromosome from a male Neandertal who lived at El Sidrón, Spain, 49,000 years ago has not been passed onto modern humans, researchers report today in The American Journal of Human Genetics. The finding fits with earlier studies that have found that although living Asians and Europeans have inherited 1% to 3% of their DNA from their ancestors’ interbreeding with Neandertals, they are missing chunks of Neandertal DNA on their Y chromosomes. This has suggested that female modern humans and male Neandertals were not fully compatible and that male Neandertals may have had problems with sperm production. The new study finds a clue to why: The El Sidrón Neandertal had mutations in three immune genes, including one that produces antigens that can elicit an immune response in pregnant women, causing them to reject and miscarry male fetuses with those genes. So even though male Neandertals and female modern humans probably hooked up more than once over the ages, they may have been unable to produce many healthy male babies (such as the reconstruction of this Neandertal boy from fossils from Gibraltar)—and, thus, hastened the extinction of Neandertals.
There are also no Neanderthal mtDNA sequences in modern humans. This is probably because the children of hybrid couples were probably raised in the tribe of the mother. Hybrid children with modern human mothers were raised in modern human tribes and left descendants who are alive today. Hybrid children with Neanderthal mothers were raised in Neanderthal tribes that ultimately went extinct. 

This inferred pattern of matrilocality for hybrid children, despite the fact that Neanderthals were predominantly patrilocal, also favors the possibility that hybrid children were largely a result of rape or brief encounters, rather than than marriage-like couple relationships. There is not a single example archaeologically of a hominin tribe containing both full blooded Neanderthal and full blooded modern human members.

Wednesday, April 11, 2018

Modern Humans At Monte Verde Ca. 18,500 Years Ago?

Two and a half years ago, a paper that I had missed announced strong evidence of a human presence in Monte Verde (in South America) about 18,500 years ago. This is about 4,000 years older than the next oldest reliably dated evidence (from the same site).

Less significantly, evidence of wild potato use in Utah (which was converted into flour) from 10,900 years ago.

Tuesday, April 10, 2018

Koreans v. Japanese v. Chinese Genetic Ancestry

A new paper purports to be able to distinguish almost perfectly between Koreans, the Japanese, and Chinese individuals based upon autosomal genetics. This is something that consumer genetic services such as 23andMe struggle with right now.

For example, 23andMe claims that my son has more Japanese ancestry than Korean ancestry, when, in fact, he probably has no recent Japanese ancestry and has entirely North Korean ancestry on his mother's side (something upon which we have very solid genealogical support from living memory back to his great-great grandparents because his last great-grandparent died just a few years ago at the age of 98).

In the case of 23andMe, this is probably because ancestry in Japanese people from Yayoi people (including mounted warriors) who invaded Japan from Korea ca. 1000 BCE is being classified as Japanese in origin, rather than Korean, despite the fact that these genetic component originated in Korea, because there are more Japanese individuals in their samples than there are Koreans. 

The paper also controversially suggests that these three populations have a common origin at about 1200 BCE (in the Shang Dynasty of China). 

Indeed, we are pretty much certain that this is not true in the case of the Japanese who have substantial Jomon ancestry that has a much greater time depth. 

The common origin hypothesis is also contrary to the hypothesis that the non-Jomon genetic component of the Koreans and Japanese has a significant "Altaic" component (also hypothesized linguistically) derived from Manchuria and Mongolia and the vicinity, in addition to substantial Chinese admixture in both Korean and Japanese populations, which is well documented historically at least in the case of the Japanese.

So, this 1200 BCE common origin date may be due to some methodological artifact (i.e. flaw) that might, perhaps, simply be measuring a common origin for the Chinese admixture component in all three populations that is significant in all of them, since this is around the time that the Chinese component of these nations would have diverged historically.

Both the linguistic and genetic origins of the Yayoi prior to their migration from Korea to Japan are disputed. Wikipedia (from the link above) notes that:
There are several hypotheses about the origin of the Yayoi people. 
The most popular one is that they were the people who brought wet rice cultivation to Japan from the Korean peninsula and Jiangnan near the Yangtze River Delta in China.[1] This is supported by archeological researches and bones found in today southeastern China.[2] 
Another is that they are from Primorskaya Oblast or northern part of the Korean peninsula. It is because the human bones of the Doihigahama ruins resemble the ancient human bones of the northern part of the Korean peninsula, and pottery is similar to the "Engraved band sentence pottery", widely used during the Yayoi period, was discovered in the Sini-Gai culture in the southwestern coastal province of Primorskaya Oblast.[3] 
The theory that Yayoi people have multiple origins is suggested and influential.[4][5] 
It is estimated that Yayoi people mainly belonged to [Ed. Y-DNAHaplogroup O-M176 and Haplogroup O-M122 which are typical for East- and Southeast-Asians.[6] 
The Yayoi may have spoken an Austroasiatic language or Tai-Kadai language, based on the reconstructed Japonic terms *(z/h)ina-Ci 'rice (plant)', *koma-Ci '(hulled) rice', and *pwo 'ear of grain' which Vovin assumes to be agricultural terms of Yayoi origin. Vovin suggests that Japonic was in contact with Austronesian, before the migration from Southern China to Japan, pointing to an ultimate origin of Japonic in southern China.[7][8] Although Vovin (2014)[8]does not consider Japonic to be genetically related to Tai-Kadai, he suggests that Japonic was later in contact with Tai-Kadai, pointing to an ultimate origin of Japonic in southern China with possible genetic relation to Austroasiatic
There is typological evidence that Proto-Japonic may have been a monosyllabic, SVO syntax and isolating language; which are features that the Austroasiatic languages also famously exhibit.[8] 
A 2015 analysis using the Automated Similarity Judgment Program resulted in the Japonic languages being grouped with the Austroasiatic languages. The same analysis also showed a connection to Ainu languages, but this is possibly because of heavy influence from Japonic to Ainu.[9]
Many linguistic scholars, however, think that Japonic and Korean are distant members of the Altaic language family which also includes the Turkish, Mongolian and Manchurian (a.k.a. Tungistic) languages.

Japanese has borrowed heavily in the historic era from the Chinese language, without changing its underlying grammar and structure, and has virtually no linguistic connection to the Ainu language which was related to the language of the Jomon people who were invaded by the Yayoi. 

This lack of linguistic borrowing from the Jomon is the case even though Jomon genetic input into the modern Japanese gene pool was very substantial for a hunter-gatherer population, perhaps because hunter-gatherers who fish, like those in Japan, the Pacific Northwest and the Baltic-Finnish region, seem to have more staying power than terrestrial hunter-gatherers, vis-a-vis farmers and herders.

Rethinking The Survival Of The Flattest

In evolution, there is a tradeoff between genomes that are maximally fit but particularly vulnerable to deleterious mutations, and those that are more capable of ignoring deleterious mutations that are less maximally fit, a type of genome known as a "flat" genome. Hence, the notion "survival of the flattest". 

But, it turns out that the trade off is not a perfect one, so minimizing the load in a genome of deleterious mutations still confers improved fitness, according to a recent pre-print:
Populations exposed to a high mutation rate harbor abundant deleterious genetic variation, leading to depressed mean fitness. This reduction in mean fitness presents an opportunity for selection to restore adaptation through the evolution of mutational robustness. In extreme cases, selection for mutational robustness can lead to "flat" genotypes (with low fitness but high robustness) out-competing "fit" genotypes with high fitness but low robustness-a phenomenon known as "survival of the flattest". 
While this effect was previously explored using the digital evolution system Avida, a complete analysis of the local fitness landscapes of "fit" and "flat" genotypes has been lacking, leading to uncertainty about the genetic basis of the survival of the flattest effect.  
Here, we repeated the survival of the flattest study and analyzed the mutational neighborhoods of fit and flat genotypes. We found that flat genotypes, compared to the fit genotypes, had a reduced likelihood of deleterious mutations as well as an increased likelihood of neutral and, surprisingly, of lethal mutations. This trend holds for mutants one to four substitutions away from the wild-type sequence. We also found that flat genotypes have, on average, no epistasis between mutations, while fit genotypes have, on average, positive epistasis. 
Our results demonstrate that the genetic causes of mutational robustness on complex fitness landscapes are multifaceted. While the traditional idea of the survival of the flattest effect emphasized the evolution of increased neutrality, others have argued for increased mutational sensitivity in response to strong mutational loads. Our results show that both increased neutrality and increased lethality can lead to the evolution of mutational robustness. Furthermore, strong negative epistasis is not required for mutational sensitivity to lead to mutational robustness. Overall, these results suggest that mutational robustness is achieved by minimizing heritable deleterious variation.

A Notable Paper Potentially Falsifying String Theory

There are vast numbers of mathematically possible concrete theories called vacua within the class of theories that make up string theory. This makes finding the needle in the haystack that could match reality difficult. This has been one of the main obstacles to developing a string theory phenomenology that can be tested with experiments.

But, it turns out that all, or almost all, of the vacua are topologically inconsistent with observed reality which has a "de Sitter" topology:
De Sitter space and anti-de Sitter space are named after Willem de Sitter (1872–1934), professor of astronomy at Leiden University and director of the Leiden Observatory. Willem de Sitter and Albert Einstein worked in the 1920s in Leiden closely together on the spacetime structure of our universe. 
In the language of general relativity, de Sitter space is the maximally symmetric vacuum solution of Einstein's field equations with a positive cosmological constant (corresponding to a positive vacuum energy density and negative pressure). When n = 4 (3 space dimensions plus time), it is a cosmological model for the physical universe; see de Sitter universe
De Sitter space was also discovered, independently, and about the same time, by Tullio Levi-Civita.
The class of topologically inconsistent vacua are called the "Swampland" and if all or most String theory vacua are in the Swampland then the task of finding a String theory vacuum consistent with our reality, or ruling out String theory entirely becomes much easier.

Woit notes an important new pre-print on "Swampland" string theory vacua (paragraph breaks  and emphasis in bold added), which suggests that all of the Sting theory vacua are in the Swampland.
I haven’t paid much attention to the Swampland business since then, but noticed last night a new preprint with the title What if string theory has no de Sitter vacua?. The authors summarize their argument:
From this analysis we conclude that string theory has not made much progress on the problem of the cosmological constant during the last 15 years. There is a general agreement that the presence of dark energy should be an important clue to new physics. So far, string theory has not been up to the challenge. Or to be more precise, string theorists have not been up to the challenge. 
The well-motivated introduction of the anthropic principle and the multiverse, was a big relief. The mathematical standards were lowered, and unconstrained model building could set in exploring a wild and free landscape of infinite possibilities. But beyond this suggestive connection between a possible multiverse and the rich mathematical structures of string theory not much solid results have been achieved. 
We reviewed some fraction of the mounting evidence that most, if not all of this landscape, is a swampland and we refer to [14,16,149] for similar lines of thought. We believe it makes more sense to listen to what string theory is trying to tell us, then to try to get out of the theory what one would like to have. In recent years, especially with the program of the Swampland [14, 150–152], there is luckily a growing community that embraces this idea. Perhaps this program really already made its first prediction: no measurable tensor modes in the CMB. 
From what we have seen so far, we believe that the most sensible attitude is to accept there are no dS vacua at all because string theory conspires against dS vacua. 
The suggestion here is basically that effective field theories on a deSitter background are in the Swampland, so can’t be derived from string theory. Since we seem to live in a deSitter space, the obvious conclusion to draw from this is that string theory is falsified: it can’t be the fundamental theory we are looking for. The authors discuss various unconvincing ways to try and avoid this conclusion.
Physicists have been working on String Theory for basically my entire life. It has turned out to be a dead end and a distraction for the most part, although it has revealed some important mathematical insights including some relevant to quantum gravity. It is time to look for alternatives.

For what it is worth, it is also possible that gravity and the cosmological constant observed are not actually topological effects as in General Relativity, but are instead has a mechanism that involves the behavior of gravitons in a spacetime that is itself, fundamentally flat rather than being deSitter or anti-deSitter, even though that graviton behavior is similar to and in most circumstances almost exactly equivalent to, a topologically curved spacetime.

Monday, April 9, 2018

Ancient DNA From Patagonia Supports Paradigm

The latest ancient DNA study from South American, although the ancient DNA is just 500 years pre-Columbian, supports the existing paradigm of New World pre-historic demography. Notably, there do not seem to be any hints of Paleo-Asian DNA in these populations.
Recent genomic studies of ancient and modern humans from the Americas have given a comprehensive view of the peopling of the continent. However, regional characterization of ancient and modern individuals is lacking, being key to unveiling fine-scale differences within the continent. We present genome-wide analyses of ancient and modern individuals from South America from Western Patagonia. We found a strong affinity between modern and ancient individuals from the region, providing evidence of continuity in the region for the last ∼1,000 years and regional genetic structure within Southern South America. In particular, the analysis of these ancient genomes helps address questions related to the maritime tradition in the region and its diversification posterior to the split from terrestrial hunter-gatherers. 
Patagonia was the last region of the Americas reached by humans who entered the continent from Siberia ∼15,000–20,000 y ago. Despite recent genomic approaches to reconstruct the continental evolutionary history, regional characterization of ancient and modern genomes remains understudied. Exploring the genomic diversity within Patagonia is not just a valuable strategy to gain a better understanding of the history and diversification of human populations in the southernmost tip of the Americas, but it would also improve the representation of Native American diversity in global databases of human variation. Here, we present genome data from four modern populations from Central Southern Chile and Patagonia (n = 61) and four ancient maritime individuals from Patagonia (∼1,000 y old). Both the modern and ancient individuals studied in this work have a greater genetic affinity with other modern Native Americans than to any non-American population, showing within South America a clear structure between major geographical regions. Native Patagonian Kawéskar and Yámana showed the highest genetic affinity with the ancient individuals, indicating genetic continuity in the region during the past 1,000 y before present, together with an important agreement between the ethnic affiliation and historical distribution of both groups. Lastly, the ancient maritime individuals were genetically equidistant to a ∼200-y-old terrestrial hunter-gatherer from Tierra del Fuego, which supports a model with an initial separation of a common ancestral group to both maritime populations from a terrestrial population, with a later diversification of the maritime groups.

Modern Human Remains In Arabia 86,000 Years Ago

Tools associated with modern humans in inland Arabia have suggested that modern humans arrived there ca. 125,000 years ago, but modern human remains are much more powerful evidence and the latest find on that front is from at least 86,000 years ago.
Understanding the timing and character of the expansion of Homo sapiens out of Africa is critical for inferring the colonization and admixture processes that underpin global population history. It has been argued that dispersal out of Africa had an early phase, particularly ~130–90 thousand years ago (ka), that reached only the East Mediterranean Levant, and a later phase, ~60–50 ka, that extended across the diverse environments of Eurasia to Sahul. However, recent findings from East Asia and Sahul challenge this model. Here we show that H. sapiens was in the Arabian Peninsula before 85 ka. We describe the Al Wusta-1 (AW-1) intermediate phalanx from the site of Al Wusta in the Nefud desert, Saudi Arabia. AW-1 is the oldest directly dated fossil of our species outside Africa and the Levant. The palaeoenvironmental context of Al Wusta demonstrates that H. sapiens using Middle Palaeolithic stone tools dispersed into Arabia during a phase of increased precipitation driven by orbital forcing, in association with a primarily African fauna. A Bayesian model incorporating independent chronometric age estimates indicates a chronology for Al Wusta of ~95–86 ka, which we correlate with a humid episode in the later part of Marine Isotope Stage 5 known from various regional records. Al Wusta shows that early dispersals were more spatially and temporally extensive than previously thought. Early H. sapiens dispersals out of Africa were not limited to winter rainfall-fed Levantine Mediterranean woodlands immediately adjacent to Africa, but extended deep into the semi-arid grasslands of Arabia, facilitated by periods of enhanced monsoonal rainfall.
H.S. Groucutt et al. "Homo sapiens in Arabia by 85,000 years ago." Nature Ecology & Evolution. (April 9, 2018) doi:10.1038/s41559-018-0518-2.

Commentary by the authors of the paper notes that:
According to the old textbook view, our species evolved in Africa about 200,000 years ago. Despite a brief, failed expansion to the edge of Eurasia about 100,000 years ago when humans first tried migrating to the lands at the eastern end of the Mediterranean (the Levant), we only successfully spread out of Africa around 60,000 to 50,000 years ago. 
Recent evidence suggests that much of this narrative is wrong. Findings in Africa, such as from the site of Jebel Irhoud in Morocco, suggest that Homo sapiens appeared early, more than 300,000 years ago. Our origin does not seem to have occurred in only one small area, but across much of Africa. 
Findings from the Levant, most recently the dating of a maxilla (upper jawbone) from Misliya Cave in Israel, suggest our species repeatedly expanded into the winter-rainfall fed, forested area just outside Africa. We don’t yet know if people survived long term in the Levant, which is a very small area. It seems more likely that there were repeated migrations from Africa. 
But what about the areas beyond the Levant? Recent findings suggest that our species got to East Asia and Australia much earlier than had been thought. But determining the hominin species present and the age of these sites have proven challenging. 
Our finger fossil gives us a more specific time range to work with, which correlates with other evidence. Stone tools from Al Wusta are similar to those from the Middle Palaeolithic (Stone Age) period in the Levant and north-east Africa. They suggest that our early spread into Eurasia was not associated with some kind of technological breakthrough, such as the invention of projectile technology as some have suggested
Together, these findings show that Homo sapiens had spread beyond the Levant much earlier than traditional accounts would have it. The Al Wusta phalanx is the oldest directly dated fossil of our species beyond Africa and the Levant and so represents a crucial reference point in understanding this topic.
Here is a link to images of arrowheads found by the same investigator.

Sunday, April 8, 2018

Quote Of The Day

Of course the scientific paper will not die any time soon. I know that because I've measured the approximate speed by which academia moves, and it's about 0.01 mm per millenium.
Sabine Hossenfelder‏ @skdh Apr 5, 2018

Thursday, April 5, 2018

Something Very Amiss In Archaic Admixture Presentation

A conference presentation on archaic admixture including Denisovan admixture purports to show an admixture date of 25,000 to 40,000 years ago, which isn't remotely plausible given existing archaeological knowledge (it should be more like 50,000 to 65,000 years ago). 

For what it's worth, I do like the "heart" icons for admixture events (even though my suspicion is that those admixture events were not actually long term romantic encounters, and may have been mostly rapes).

The abstract of the paper from 2016 discussed in the presentation is as follows:
The population history of Aboriginal Australians remains largely uncharacterized. Here we generate high-coverage genomes for 83 Aboriginal Australians (speakers of Pama–Nyungan languages) and 25 Papuans from the New Guinea Highlands. We find that Papuan and Aboriginal Australian ancestors diversified 25–40 thousand years ago (kya), suggesting pre-Holocene population structure in the ancient continent of Sahul (Australia, New Guinea and Tasmania). However, all of the studied Aboriginal Australians descend from a single founding population that differentiated ~10–32 kya. We infer a population expansion in northeast Australia during the Holocene epoch (past 10,000 years) associated with limited gene flow from this region to the rest of Australia, consistent with the spread of the Pama–Nyungan languages. We estimate that Aboriginal Australians and Papuans diverged from Eurasians 51–72 kya, following a single out-of-Africa dispersal, and subsequently admixed with archaic populations. Finally, we report evidence of selection in Aboriginal Australians potentially associated with living in the desert.
Anna-Sapfo Malaspina, et al., "A genomic history of Aboriginal Australia" 538 Nature 207-214 (October 13, 2016) doi:10.1038/nature18299

More details about the sample, including uniparental haplotypes are available here. The main supplements are here. The dating is discussed in the S07 supplemental materials.

No subsamples of Aboriginal Australians (except three outlier individuals two of whom have known mixed Papuan and Australian ancestry) have greater than average genetic affinities to any particular Papuan population and no Papuans in the sample have greater than average genetic affinities to any particular Australian population. There is no significant difference in Denisovan admixture proportions between Papuans and Aboriginal Australians after controlling for recent Eurasian admixture. ANCESTRY software analysis supports a "shared genetic origin of Aboriginal Australians, Papuans and Bougainvilleans[.]" Various scenarios for one or two Out of Africa events and one or two Neanderthal admixture events, in addition to one Denisovan admixture event in all cases are nearly degenerate in their fit to the data.

The Denisovan related population that admixes with Papuans and Australian Aborigines is about 400,000 years diverged from the Altai Denisovans per S07 supplemental materials.

All Papuan and Australian populations in the study as well as the mainland Asian Dai population picked as representative, show a bottleneck in effective population size to 2,000 to 3,000 people from about 60,000 years ago to 40,000 years ago, preceded by falling effective population size back to the dawn of the species, followed by growing effective population sizes. But, several Australian sub-populations show a dip in effective population size from 10,000 years ago to 5,000 years ago.

Given the discussion in the abstract, it would be possible to reconcile the genetic based estimates and the archaeology if the archaic admixture actually took place coincident with divergence from Eurasians 51-72 kya. In that scenario, the 25-40 thousand years ago date could reflect the point at which Papuan and Aboriginal Australians ceased to be a common gene pool in Sahul because they ceased to exchange brides and/or husband with each other, perhaps as the last major ice age began to have an impact. A bottleneck event in Australia later (perhaps around the time of the Last Glacial Maximum or the Younger Dryas), or at about the same time, wiping out all other Australian populations and reducing the last one to a small founding population could then have occurred ca. 10-32 kya.

I remain skeptical, however, because other genetic studies have pointed to an almost immediate divergence of Papuans from Aboriginal Australians at about the same time as the two populations crossed the Wallace line and diverged from other Eurasians. Also, the archaeology suggests that there were little trade or interaction between the Papuans and the Aboriginal Australians, even during the time period when they could have interacted over a land bridge.

This Wikipedia summary regarding the ancient contours of Sahul states (omitting references to much later Austronesian contact in the Holocene era, i.e. less than 10,000 years ago):
Humans first populated eastern Wallacea (including Timor, which at the time was separated from mainland Sahul), the rest of Sahul, and the Bismarck Archipelago from Sunda around 45,000 years ago, by a founding population estimated to have been at least several hundred, and having had relatively sophisticated water craft.[44] There was little subsequent population mixing between Wallacea and Sahul for about 30,000 years, and indeed relatively little mixing between the north and south and the east and west of Sahul after the initial dispersal of the population. 
Indigenous Australians are the original inhabitants of the Australian continent and nearby islands who migrated from Africa to Asia around 70,000 years ago[45] and arrived in Australia around 50,000 years ago.[46] They are believed to be among the earliest human migrations out of Africa. . . .
The original inhabitants of the group of islands now named Melanesia were likely the ancestors of the present-day Papuan-speaking people. Migrating from South-East Asia, they appear to have occupied these islands as far east as the main islands in the Solomon Islands, including Makira and possibly the smaller islands farther to the east.[48]
Also, the estimate for arrival of humans in Australia is conservative. This linked Wikipedia article notes evidence of humans in Australia 65,000 years ago citing to: Clarkson, Chris, et al., "Human occupation of northern Australia by 65,000 years ago" 547 (7663) Nature 306–310 (July 19, 2017) doi:10.1038/nature22968.

A minority view that could explain the discrepancy between the genetic data and the archaeological data (which is admittedly thin before about 40,000 years before present), is that the first wave of modern humans in Australia could have gone extinct or become moribund only to be replaced by a second wave of settlers from Wallacea or Papua New Guinea that was also Denisovan admixed. As one Wikipedia footnote explains:
The dominant view among archaeologists and palaeoanthropologists is that the earliest humans in Australia were among the ancestors of modern Aborigines. There is some evidence to the contrary. For instance, among the Lake Mungo remains, the mtDNA of the remains known as LM3 do not indicate a close relationship to modern Aborigines, according to "Mitochondrial DNA sequences in ancient Australians: Implications for modern human origins" (PNAS, volume 98, issue 2) by Adcock et al. This finding has not been widely accepted by the palaeoanthropological community.
There are certainly examples of first wave colonization efforts of new territory that failed (e.g., the Leif Erikson's Vikings in North America ca. 1000 CE, and Roanoke in what would become the United States, and Madagascar in the early 1st Millenium), only to be followed by colonizations that thrived later on, however. And, there are also examples of near total population replacement (e.g. Great Britain in the Bell Beaker era) in a place. So, this theory can't be entirely dismissed out of hand, so long as the resettling population also had Denisovan admixture.

Another possibility is that there is a problem with the methodology or assumptions of the 2016 paper. For example, that papers could be using inaccurate assumptions about effective population size or population structure or mutation rates or generation duration that distort the paper's estimates.