The PCA method is flawed according to a study

In a recent post I discussed a study that questioned the validity of D-statistics; today I will explore two research papers that look into PCA or Principal Components Analysis, a statistical tool widely used in genetics and anthropology, which, according to two papers, may be biased, subjective, and flawed.

What is PCA?

Principal Components Analysis or PCA is a tool used to analyze data. It can be simplified as a method that concentrates, or reduces this data into two dimensions, where the data can be plotted and patterns, or structure identified. Points closer to one another should be similar and associated, those that are more distant in the plot are assumed to be more weakly (or not at all) linked.

You can obtain different information from samples taken from a population, and if there are, say 15 different variables measured for a given sample, it is very complex to pair up the variables and compare the samples to reach a conclusion.

Suppose you are studying bacteria, both pathogens and benign ones, and have defined fifteen variables, collecting data for each of them (length, width, thickness, rugosity, motility, lifespan, etc.). Your goal is to evaluate the influence of these factors on bacterial virulence. It would be impossible unless you could somehow simplify the data without losing information. Imagine if you could concentrate those 15 variables (or 15 dimensions) into just two, it would be much easier to visualize differences or similarities for your analysis.

PCA does just that. Using statistical formulas and computer power it condenses these variables into two main ones, "Principal Components" PC1 and PC2 without losing information, and plots them in a 2-dimensional chart.

Below is a simplified graph showing how three variables, that form a three dimensional space can be compressed into two Principal Components. The points (one sample is equal to one point), are graphed into a three dimensional space, with the three variables, x, y, and z defining their position.

As you can see, the first line (blue) is drawn along the region with most variability in the samples. A second line is drawn perpendicular to it (red). The First Principal Component is PC-1, the other is the Second Principal Component or PC-2. The data is theb "transformed" into a new, 2-D chart, where each point is one of the original samples from the original dataset.

I won't go into statistic, but the transformation is based on evaluating how the variance of the variables of the dataset (how "far" they are from the mean value of the total set), and how they change together (covariance). An example of covariance could be: alchol intake in oz. and driving ability (the higher the intake, the lower the ability to drive).

This analysis results in one main Component PC-1, which reproduces the highest variance, followed by all other perpendicular components ranked by importance. Of these, the second (PC-2) component is chosen.

We should take into account that since PC-1 is more important than PC-2 the differences along the PC-1 axis are greater than those along the PC-2 axis even if the "distance" between points is the same.

Below is a real example from a paper on human population genetics.

PCA global populations. Adapted from Geleta et al, Deep Variational Autoencoders for Population Genetics September 2023 doi:10.1101/2023.09.27.558320

The Africans seem to be most distant from all other groups and compact (more similar with each other), while West Asians are widely spread out along PC-1 (showing intra-population differences); Oceanians seem to form another compact group.

The method is faulty

Thhhe first article criticizing this PCA method is Principal Components Analysis fails to recover phylogenetic structure in hominins, by Levi Y. Raskin, et al., bioRxiv doi: https://doi.org/10.1101/2025.10.31.685754 🔓 (it has been published now in the American Journal of Biological Anthropology doi: 10.1002/ajpa.70306; Vol 190, no. 3: e70306. https://doi.org/10.1002/ajpa.70306.🔒). The authors analyzed studies using "morphometric data" (lengths, widths, angles, distances, thicknesses of bones, skulls, specimens) and found that "...PCA trees inferred from traditional morphometric data were identical to the sampled tree in 0.11% of datasets when we only considered PC axes 1 and 2, and in 2.9% of datasets when we considered all axes. No PCA tree inferred from any of the 2,400,000 shape datasets was identical to the sampled tree, regardless of the number of axes."

Only 11 out of 1000 trees produced by PCA methods with 2 dimensions were valid! And using more dimensions, the success ratio was less than three percent!

This questions the validity of conclusions reached using PCA methods: "Phylogenetic interpretations of the hominin fossil record based on proximity in PC space are inherently flawed and likely to be erroneous. Arguments in the hominin systematics literature based on PCA should therefore be reevaluated using phylogenetically-informed alternatives."

The authors are terminant. Those were best case scenarios. In real studies, bias and missing data could make matters even worse. The paper concludes that "PCA is fundamentally ill-suited and unreliable for inferring hominin evolutionary relationships from morphological data. Studies of hominin systematics that incorporate, and especially those that rely on, PCA as a means of inferring morphological and phylogenetic affinity should therefore be revisited."

This is relevant because many PCA charts using morphometric data assume that those hominins who are closer to each other in the chart, are related in the family tree and belong to a similar lineage. But it seems that this may not be the case. Below is an example of a morphometric PCA (Fig. 6 in Source); in the caption the text says "... High-scoring test subjects cluster together, suggesting cranial morphological similarity between them...":

Another Critical Voice: PCA and its use in genetic studies

An earlier paper published Aug. 29, 2022 by Elhaik E. Principal Component Analyses (PCA)-based findings in population genetic studies are highly biased and must be reevaluated (Sci Rep. 2022; 12(1):14683. doi: 10.1038/s41598-022-14395-4. PMID: 36038559; PMCID: PMC9424212), questioned the validity of conclusions obtained by using PCA. Elhaik is highly critical about its applicability to genetic studies and suggests that they are unreliable, and lead to false interpretations.

"...We demonstrate that PCA results can be artifacts of the data and can be easily manipulated to generate desired outcomes. PCA adjustment also yielded unfavorable outcomes in association studies. PCA results may not be reliable, robust, or replicable as the field assumes. Our findings raise concerns about the validity of results reported in the population genetics literature and related fields that place a disproportionate reliance upon PCA outcomes and the insights derived from them. We conclude that PCA may have a biasing role in genetic investigations and that 32,000-216,000 genetic studies should be reevaluated."

This work looks into the population PCAs, like the one shown furhter up (Africans, Europeans, Asians, etc). It notes that the PCA charts place the populations along the corners of an imaginary triangle: Africans, Europeans, and East Asians on each vertex of the triangle. And this is assumed to be the result of the migration out of Africa, followed by separate divergence and mutations. A conclusion widely accepted by geneticists. However, Elhaik says it isn't so: "Here we show that the appearance of continental populations at the corners of a triangle is an artifact of the sampling scheme since variable sample sizes can easily create alternative results as well as alternative “clines”."

This study states that "PCA can be used to generate conflicting and absurd scenarios, all mathematically correct but, obviously, biologically incorrect and cherry-pick the most favorable solution. This is an example of how vital a priori knowledge is to PCA. It is thereby misleading to present one or a handful of PC plots without acknowledging the existence of many other solutions, let alone while not disclosing the proportion of explained variance."

The manipulation implied by the author is that the outcome of the PCA chart is controlled by the scholar who already "knows" what the chart should reflect "By manipulating the choice of populations, sample sizes, and markers, experimenters can create multiple conflicting scenarios with real or imaginary historical interpretations, cherry-pick the one they like, and adopt circular reasoning to argue that PCA results support their explanation".

Closing comments

PCA is a valid statistical tool, with ample use in marketing, segmentation of customers, in facial recognition, in finance, climate modelling, medicine, and engineering (predictive maintenance, quality control). But, it seems that it use in the fields of anthropology and genetics is not robust or reliable.

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2026 by Austin Whittall © 

Bitter Taste genes

The ability to taste bitterness may help us avoid poisoning from eating toxixc plants. Animals in general, and our ancestors, the primates developed this trait. The PTC gene is responsible for our perception of bitter taste.

All nonhuman primates have only one variant of the PTC gene known as PAV, which is therefore considered as the ancestral or original gene form, these apes are homozygous for it (meaning that the two copies they carry, one from each parent, are identical). This is known as the "taster" allele, it allows them to taste bitterness.

Genetics

PTC bitter gene, is formally known as the TAS2R38 gene. It comes in eight different variants but two of them are prevalent, the ancestral "taster" allele, and a "non-taster" allele which comprise 96% of the human population. These encode a specific protein, which contains 333 aminoacids, the 7-transmembrane domain G-protein-coupled receptor which responds to bitterness.

The different alleles cause tiny variations in the position of some aminoacids in this protein and cause the "taster" and "non-taster" variants and four other intermedieate "less-taster" types.

The variants are named after the positions of these amino acids the "original" or "ancestral" form is the PAV form (because it contains proline at position 49, alanine at position 262, and valine at position 296), this is the "taster" form.

Neanderthals and Denisovans are also PAV tasters (Source).

The second mayor form is the "non-taster" one, known as AVI, because it contains alanine, valine and isoleucine aminoacids in those three positions, respectively.

Further down we will look into why do the "non-taster" alleles survive, and account for roughly half of the human population, who can't taste bitter flavors. If tasting bitterness protects against plant toxins, why do so many of us carry the non-taster variant?

The other six variants are AAV, AVV, AAI, PAI, PVI, AAI, and PVV and are found at relatively low frequencies.

The prevalence found in one study was the following: 42.3% PAV (ancestral, taster), 53.1% (derived, non-taster) and the intermediate taster ones (2.5% AAV, 1.2% AAI, 0.8% PAI, and 0.1% PVI, no AVV or PVV were detected). (Source).

People who inherited at least one copy of the PAV allele from their parents are able to taste bitterness.

These alleles also have a geographic distribution, PAV and AVI are the most frequent, and make up the vast majority of European and Asian alleles. They are also found in Africa, but there, AAI is found at a relatively high frequency. The table below, (Table 1 from Risso, D., Mezzavilla, M., Pagani, L. et al. (2016) Global diversity in the TAS2R38 bitter taste receptor: revisiting a classic evolutionary PROPosal. Sci Rep 6, 25506), shows data from 5,589 individuals sampled across 105 populations around the World, and it highlights the slight variations in different populations.

Table 1: Detailed distributions of TAS2R38 haplotypes in the studied populations. . Source

The "Americans" in the table shown in the image are from the following groups (the number is the individuals in each sample): North America Maya Mexico 42, North America Puerto Ricans Puerto Rico 110, South America Colombians Colombia 134, South America Karitiana Brazil 28, South America Mexicans Mexico 160, South America PEL Peru 170, and South America Surui Brazil 16.

American Natives and Bitter Taste Genes

Looking at the data, we see that PVV is exclusively European, and does not appear in any other population, including Amerindians. It didn't admix into them despite the large-scale intermingling that took place after 1492, which is quite surprising.

AAI is definitively African where it reaches 13.22%, and in small amounts among Europeans and Americans (perhaps due to African genetic mixing into Southern Europeans and slave trade into America).

PVI is extremely rare, and is found at higher frequencies among Native Americans, with 0.19%, followed by Africans by 0.15%. None in Asia, and only 0.03% in Europe. We could suppose that African slave trade brought it into America, but why is the prevalence 26.6% higher in the Americas than in Africa? Being absent in Asia it surely didn't arrive via Beringian migrants.

AAV, absent in Asians is also found among Americans (2.26%), slightly lower than Europeans (3.56%) and much higher than Africans (0.61%). If it introgressed into Amerindians through Europeans, then, why didn't the European AVI do so in a similar proportion? (AVI among Americans is 26.69% while it is 49.22% among Europeans and roughly 33% in Africans and Asians).

Regarding the ancestral PAV, original allele, it is highest among Native Americans with 68.8%. The other populations have a lower frequency of it.

Interestingly, according to Kim et al. (See: Kim UK, Jorgenson E, Coon H, Leppert M, Risch N, Drayna D., Science. 2003 Feb 21;299(5610):1221-5. doi: 10.1126/science.1080190. PMID: 12595690. Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide), "The common nontaster AVI haplotype was observed in all populations except Southwest Native Americans, who were exclusively homozygous for the PAV haplotype" (these natvies were almost 100% tasters).

Another article (Flores SV, Roco-Videla A, Aguilera-Eguía R. Variation in haplotype frequencies of the TAS2R38 gene, associated with the perception of bitter taste. Salud, Ciencia y Tecnología. 2025 Jan. 1;5:1026.) notes taster prevalence among Peruvian Andean people: "A particularly interesting case is the Peruvian population, which stands out for its high frequency of bitter taste perception diplotypes. In this population, only 1 % has the AVI/AVI diplotype, indicating an almost total prevalence of bitter taste perception (PAV/PAV and PAV/AVI). This exception suggests a specific dietary adaptation in the Andean region, or well the result of genetic drift."

Why hasn't natural selection erased the non-taster alleles?

For human beings, nearly all naturally occurring plant toxins poisons taste bitter. But, not all bitter tasting foods are poisonous. Many bitter tasting foods are harmless.

One interesting paper suggests that excluding all bitter flavored plants would mean lost calories and nutrients, because many bitter veggies are healthy and have no harmful effect (citric fruits, bitter melon, or kale, as well as the other cruciferous vegetables), some foods like beer, green tea, or coffee are bitter yet pleasurable. Also, bitterness may also mean medical properties such as quinine (the bitter ingredient of tonic water) used to combat malaria, or the pain-killing properties of salicin, found in willow leaves, on which the aspirin was based.

Humans also have cognition, and curiosity, they may try a bitter food, which if it doesn't cause harm, can then be safely added to the diet.

A paper (T2R38 taste receptor polymorphisms underlie susceptibility to upper respiratory infection Robert J. Lee,… , Danielle R. Reed, Noam A. Cohen. Published October 8, 2012. Citation Information: J Clin Invest. 2012;122(11):4145-4159. https://doi.org/10.1172/JCI64240.) found that bitter taste receptors also act upon the tissue lining the upper respiratory tract, and those carrying at both PAV alleles (tasters) are better protected from microbes than those carrying one or none: "these individuals are more likely to be infected with gram-negative bacteria such as P. aeruginosa than those with 2 functional receptor alleles." It also suggests that " humans with more non-taster alleles live in the colder climates, where the evolutionary pressure for the taster genotype may be relaxed, as there are fewer pathogens than in warmer climates."

But why the AVI non-taster allele is still carried at around 50% levels in humans. Could it be functional for other reasons and offer an evolutionary advantage that we have not yet identified?

A 2012 paper (Campbell MC, Ranciaro A, Froment A, Hirbo J, Omar S, Bodo JM, Nyambo T, Lema G, Zinshteyn D, Drayna D, Breslin PA, Tishkoff SA. Evolution of functionally diverse alleles associated with PTC bitter taste sensitivity in Africa. Mol Biol Evol. 2012 Apr;29(4):1141-53. doi: 10.1093/molbev/msr293. Epub 2011 Nov 29. PMID: 22130969; PMCID: PMC3341826) says it does, but isn't yet fully understood:

"the selective force maintaining common AAV, AAI, and AVI haplotypes for extraordinarily long periods of time remains unclear. Although both AAV and AAI are associated with intermediate bitter taste sensitivity, the AAI haplotype is more common in Africa than AAV. Intriguingly, the AAV haplotype may represent a “stepping stone” to other more advantageous haplotype variation, such as AAI and AVI. We suggest that common PAV, AAI, and AVI haplotype variation may be maintained at high frequencies in response to selective pressures unrelated to diet. Indeed, recent studies have shown that bitter taste receptors are expressed in a variety of cell types in the human gastrointestinal tract (Rozengurt and Sternini 2007) and lungs (Shah et al. 2009; Deshpande et al. 2010), where they influence insulin and glucose levels (Dotson et al. 2008), eliminate harmful inhaled substances (Shah et al. 2009), and stimulate the relaxation of airways for improved breathing (Deshpande et al. 2010). These studies demonstrate that bitter taste loci have a number of different functions and raise the possibility that common variants at TAS2R38 may be under selection due to their physiological roles in human health beyond oral gustatory function. Though we cannot conclusively distinguish the selective forces maintaining common variation at TAS2R38, it is clear that genetic variation at this locus is diverse and has been functionally important long before modern Homo sapiens existed."

Neanderthals

Our relatives, the Neanderthals had bitter taste perception (source) the El Sidrón individual, was heterozygous, carrying the ancestral PAV and the derived allele with an alanine in position 49, the study didn't clarify the other two positions, so this Neanderthal could have carried the common non-taster AVI, or the more rare variants AAI or AAV.

This means that the non-taster variant dates to before humans and Neanderthals split around 500,000 years ago. Of course, genetic flow between both groups could have introduced the derived non-taster variant into the 48,000-year-old Sidrón individual (meaning it originated among our H. sapiens), but the study considers this unlikely and affirms that "our results indicate that the non-taster alleles were already present in the ancestral human populations from which both Neanderthals and modern humans diverged."

The rise of non-tasters took place long ago. Studying African populations Tishkoff et al. (2012) found the following evolution and timeline for this gene from the PVA to the AVI form:

The PAV → AAV variant arose when P was replaced by A at site 49 1.3 million ± 242,211 years ago. Then the AAV → AAI shift took place when V was replaced with I at position 296, 1.0 million ± 267,268 years ago. These changes predate the split between Neanderthal-Denisovans and our ancestral H. sapiens lineage. Another mutation was the A for V at position 262, causing the AAI → AVI shift. This one took place 336,000 ± 89,845 years ago. The other low-frequency alleles are much younger than 200,000 years. Below is and years old, respectively. The lower frequency variants, including those that are associated with decreased PTC sensitivity, appear to be much younger in age, occurring within the last 200,000 years. Below is Fig. 4, from this paper, we added in red letters, each allele.

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2025 by Austin Whittall © 

On the bias and weaknesses of DNA analysis of ancient genomes

I came across a very interesting article (Lumila Paula Menéndez, (2025), The antiquity and ancestral origin of humans in the americas: a five hundred year inquiry from a biological anthropology perspective. Journal of Anthropological Sciences Istituto Italiano di Antropologia, Review Vol. 102 (2024), pp. 7-67, doi7p.10.4436/jass10201) that summarizes all the theories about the peopling of America over the past centuries.

It makes good reading. In particular, it points out the shortcomings of scientific papers about this subject due to different sources of bias, lack of research, and the inconsistencies that can arise from DNA studies based on the "settings" or "initial parameters" defined by each author when using DNA analitical tools and software, a concept that I agree with 100%.

"It is essential to acknowledge that the results we have today are not without bias and may not fully represent the past. Some biases are intrinsic due to the preservation of old samples, while others are methodological, associated with different methods used, and there are geographic and geopolitical biases. Disciplinary bias arises when, for example, only cultural evidence is discussed without incorporating human biological data (and vice-versa). Geographic sampling bias occurs when, despite continental coverage in some studies, samples from South America and Central America are typically underrepresented. Studies often include a few early Holocene individuals to infer biological relationships among populations or test dispersion models (Hubbe et al. 2009; Galland and Friess 2016; von Cramon-Taubadel et al. 2017). Due to this geographical bias, archaeological sites from Latin American countries are not frequently mentioned in the literature (or the same few ones are cited over and over), perpetuating geopolitical imbalances between researchers coming from central and peripheral countries (Yañez et al. 2023). Moreover, early Holocene individuals in these studies often derive from the same few archaeological localities, potentially underestimating the wide regional variation during that period. This sample bias may impact the perception of early Holocene variation due to incomplete sampling of a highly structured population across time. After the surge of aDNA studies, some misconceptions have emerged, such as the expectation that genetic results can answer more questions about human history than realistically possible. Contrary to common belief, interpretations based on genetic results should be approached catiously. These interpretations often rely on a small number of poorly preserved samples (i.e., aDNA molecules being usually short and damaged), constituting less than 1% of Native Americans’ variation. Additionally, various decisions made during the workflow, including molecular clock calibration, reference genomes used for sequence alignment, genotype imputation, and the selection of outgroups, significantly impact the obtained results (Axelsson et al. 2008; Orlando et al. 2021). While genetic data, in combination with other information sources like archaeology can contribute to models determining approximate population divergence (Raff 2022), it is crucial to recognize that genetic results can be misleading (Steeves 2023)..."

Confirmation Bias. Copyright © 2025 by Austin Whittall

We all believe what we want to believe and ignore the facts that we don't like.

The Ancient DNA from Mengzi Ren

This is an example of how DNA data lead to wrong conclusions. This case went from proving Amerindians were linked to people living in Yunnan 14,000 years ago, to contamination by modern Han Chinese who manipulated the ancient remains! It took 3 years for someone to set the record straight.

A paper published in 2022 found affinities between the 14.000-year-old remains of a woman from Southern China, and Native Americans ( Zhang, X., Ji, X., Li, C., Yang, T., Huang, J., Zhao, Y., Wu, Y., Ma, S., Pang, Y., Huang, Y., et al. (2022). A Late Pleistocene human genome from Southwest China. Curr Biol 32, 3095–3109 e3095. doi:10.1016/j.cub.2022.06.016.)

The authors found that "14,000-years-ago human remains (Mengzi Ren, MZR) unearthed in Southwest China. MZR represents an early diversified human lineage in eastern Asia where they detect a clear genetic stratification of ancient populations. MZR deeply links to the East Asian ancestry that contributed to First Americans."

However, earlier this year, it was later found to be based on dubious data (Tabin, D., Patterson, N., Mah, M., and Reich, D. (2025). Concerns about ancient DNA sequences reported from a Late Pleistocene individual from Southeast Asia. Current Biology 35, R212–R213. doi:10.1016/j.cub.2024.10.012).

A final critique of the original paper was published by the same authors shortly after, and reported that the "ancient" DNA "data have specific affinity to modern Han Chinese and other related populations. This is a priori surprising given the date and location of the MZR individual. This affinity is so strong that MZR can be well modeled as entirely Han with added noise. Neither of these traits are shared with other ancient Southeast Asians and both raise additional concerns regarding the reliability of the MZR data. Contamination seems more plausible than a population with Han-like ancestry existing in Yunnan province 14 thousand years ago." (Tabin D, Patterson N, Mah M, Reich D. Addendum to Ancient DNA data from Mengzi Ren, a Late Pleistocene individual from Southeast Asia, cannot be reliably used in population genetic analysis. bioRxiv. 2025 Mar 26:2025.03.24.645126. doi: 10.1101/2025.03.24.645126. PMID: 40196503; PMCID: PMC11974788.).

Ancient DNA analysis involves complex procedures, and each step of the process, from extracting the sample, to sequencing it can affect the accuracy of the outcome. The following article: Anastasia V. Poznyak, Tatyana Vladimirovna Kirichenko et al., (2024). Ancient DNA studies: Common limitations and Genotyping, Journal of Angiotherapy, 8(6), 1-8, 9750, gives an enlightening overview of the process and its risks.

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2025 by Austin Whittall © 

The Genetics of Southern Patagonia

The genetics of the Southern Patagonian natives is more complex than expected. A paper published in Nature on Aug. 3, 2020 (Nakatsuka, N., Luisi, P., Motti, J.M.B. et al. Ancient genomes in South Patagonia reveal population movements associated with technological shifts and geography. Nat Commun 11, 3868 (2020). https://doi.org/10.1038/s41467-020-17656-w) explains the possible admixtures that led to the native populations in Tierra del Fuego and the Southern continental tip of Patagonia.

The authors noted that the natives that adapted to a seafaring life (marine adaptation) either "adopted the technology or invented it independently", they didn't find support that a there was a "large-scale immigration into South Patagonia of people from the north" who were already skilled in living off the sea.

Instead they found that a group of people came from the Northwest (along the Chilean Pacific Ocean coast) after the initial colonization event, around 5000 years ago (as characterized by the remains found at Ayayema ~4700 BP) these people had adopted a maritime economy and replaced the original people that had settled in this area. These latecomers:

"replac[ed] the lineages related to Punta Santa Ana (~6600 bp) and La Arcillosa2 (~5800 bp) that were previously established in South Patagonia itself. The arrival of this new stream of people could be related to the change in lithic technology between ~5500 and 3100 bp, characterized by the interruption of green obsidian use and the introduction of large biface projectile points in the Western Archipelago and Beagle Channel regions."

Archaeologists had noticed that green obsidian, a sharp volcanic glass that had been used in this region 6700 to 6300 years BP vanished, and later (5500 to 3100 BP) stone tools were made from other materials.

There was a simultaneous change in stone tools at that time.

The authors point out that "The disruption in green obsidian use has been hypothesized to reflect a loss of cultural knowledge about the location of the source of this raw material, potentially due to arrival of new people unfamiliar with the landscape."

There was another migration from the Chilean region, characterized by remains found in Central Chile. (Perhaps these people moved pushed by the migration that brought the Mapuche people into Chile?). This final group brought new genetic traits and spread between 4700 and 2000 BP. The authors wonder if the "shared linguistic family between North, Central, and South Patagonia groups in historical and modern times could be related to this signal." I assume they refer to the language of the tehuelche people, known as chon

Mapuche: Late Arrival

I have posted about the Mapuche-Guaraní link in the past. Now I have included more information on this matter in my book's second edition, a brief snippet is shown below:

Dillehay in 2007,(1) noting the similar cultural traits (ceramic, funeral practices), language similarities, and genetics, suggested an association between the Mapuche and the Amazonian natives.
A study published in 2022 analyzed ancient human DNA from the cement attaching head lice nits and found that they support a “major migration wave, occurring around 2,000 years BP from the ancient Proto Tupí/Tupí of the Amazonian lowlands (northwest Amazon) toward south Amazonia and the east Coast, spanning ∼4,000 km.”(2) The expansion continued across Argentina and reached the Andes.
This migration was probably caused by a growing population requiring new lands for farming or, as a response to a drier climate at that time. The authors concluded that “The three Calingasta [San Juan province, Argentina] ancient mummies analyzed in this study show for the first time that the original peoples of San Juan, Calingasta, came from Amazonia about 2,000 years BP… help[ing] decipher a piece of the puzzle of pre-Columbian ancient migrations within South America, of an impressive move spanning 5,000-6,500 km. [3,100-4,040 mi.]” Interestingly, Calingasta is only 160 km (100 mi.) to the east of the Mapuche homeland in Central Chile. Another study from 2023 pointed out that “the fact that together, regional varieties of Mapudungun [Mapuche language] still form a coherent single language is consistent with a relatively recent common origin, followed by geographical expansion and divergence over a timescale of the order of many centuries, but not millennia.”(3)

(1) Pedersen MW, Antunes C, De Cahsan B, et al., (2022). Ancient Human Genomes and Environmental DNA from the Cement Attaching 2,000-Year-Old Head Lice Nits. Molecular Biology and Evolution, Vol. 39, Issue 2, February 2022, msab351.
(2) Arango-Isaza E., Capodiferro MR., et al., (2023), The genetic history of the Southern Andes from present-day Mapuche ancestry. Current Biology, Vol 33:13, pp. 2602-2615.e5.
(3) Dillehay T.D. et al., (2007). Cultivated wetlands and emerging complexity in south-central Chile and long distance effects of climate change. Antiquity. 949–6.

The timing of this Out-Of-Amazon migration seems to coincide with the displacement of people into Patagonia from Central Chile

My next post will look into the Kofkeche myth, and its relation to the arrival of the Mapuche in Chile.

CAPTION HERE. Copyright © 2025 by Austin Whittall

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2025by Austin Whittall © 

Gibraltar Neanderthal DNA sequenced: they are not the last survivors.

The DNA of two Neanderthal skulls discovered in Gibraltar has revealed their sex and their affinity with the "Older" Neanderthals such as Scladina Belgium and Russia (see my recent post on the "Old" and "New" Neanderthals) instead of more recent Neanderthals from Spain.

One skull, found in 1848 in Forbes' quarry, lacks the stratigraphic information to be able to date it correctly, and it was assumed to be recent, as Gibraltar is the southernmost part of the Iberian peninsula, and therefore the last bastion where Neanderthals managed to survive when modern Homo sapiens occupied their European range.

But, according to the paper (A genetic analysis of the Gibraltar Neanderthals, Lukas Bokelmann et al. PNAS first published July 15, 2019 https://doi.org/10.1073/pnas.1903984116), this skull (also known as Gibraltar 1) is actually closer to the earlier Neanderthals than to the more recent ones:

"is genetically more similar to the ∼120,000-y-old Neanderthals from Scladina Cave in Belgium (Scladina I-4A) and Hohlenstein-Stadel Cave in Germany, as well as to a ∼60,000- to 70,000-y-old Neanderthal from Russia (Mezmaiskaya 1), than to a ∼49,000-y-old Neanderthal from El Sidrón (El Sidrón 1253) in northern Spain ".

The second skull belonged to a child, discovered in 1926. The study proved that he was a boy, but his DNA affinity was not mentioned.

DNA Tree Neanderthals. From the paper

As we can see there is "affinity" with the Scladina, Altai and HST branch but we can also see that it is also on the branch that led to all later Neanderthals.

What is really interesting is that these skulls had been handled by many people who left their human DNA on them (back in 1926 or THE 1800s nobody even dreamed of DNA) yet the scientists managed to remove it and focus only on the Neanderthal DNA. Also they found it in bones from a warm setting -heat tends to destroy ancient DNA. So that means that we can harbor hopes that DNA from hot settings in China or Africa can also be sequenced.

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2019 by Austin Whittall © 

Denisovans. More information on their genetics

A paper by S. Sawyer et al., Nuclear and mitochondrial DNA sequences from two Denisovan individuals, PNAS, doi: 10.1073/pnas.1519905112 Published Nov. 16, 2015 has sequenced the nuclear and mtDNA of two Denisovans. Their findings are the following:

Abstract
Denisovans, a sister group of Neandertals, have been described on the basis of a nuclear genome sequence from a finger phalanx (Denisova 3) found in Denisova Cave in the Altai Mountains. The only other Denisovan specimen described to date is a molar (Denisova 4) found at the same site. This tooth carries a mtDNA sequence similar to that of Denisova 3. Here we present nuclear DNA sequences from Denisova 4 and a morphological description, as well as mitochondrial and nuclear DNA sequence data, from another molar (Denisova 8) found in Denisova Cave in 2010. This new molar is similar to Denisova 4 in being very large and lacking traits typical of Neandertals and modern humans. Nuclear DNA sequences from the two molars form a clade with Denisova 3. The mtDNA of Denisova 8 is more diverged and has accumulated fewer substitutions than the mtDNAs of the other two specimens, suggesting Denisovans were present in the region over an extended period. The nuclear DNA sequence diversity among the three Denisovans is comparable to that among six Neandertals, but lower than that among present-day humans.

Take note of how large the Denisovan teeth are. From the paper

The paper notes that "Both Denisova 8 and Denisova 4 are very large compared with Neandertal and early modern human molars, and Denisova 8 is even larger than Denisova 4. Only two Late Pleistocene third molars are comparable in size: those of the inferred early Upper Paleolithic modern human Oase 2 in Romania and those of ObiRakhmat 1 in Uzbekistan"

The teeth though primitive looking also seem to differ from H. erectus teeth too. Which is quite interesting.

Denisova 8 is about 60,000 years older than Denisova 3 and Denisova 4 meaning that they are 110,000 years old. This is a long period of time in a same location.

The paper says the following about this long occupation: ", suggests Denisovans were present in the area at least twice, and possibly over a long time, perhaps interrupted by Neandertal occupation or occupations. Denisovans may therefore have been present in southern Siberia over an extended period. Alternatively, they may have been present in neighboring regions, from where they may have periodically extended their range to the Altai.".

The paper ends with an open question: " Given that the high-coverage genome from the Denisovan 3 phalanx carries a component derived from an unknown hominin who diverged 1–4 million years ago from the lineage leading to Neandertals, Denisovans, and present-day humans, it is possible that this component differs among the three Denisovan individuals. In particular, it may be that the older Denisovan population living in the cave carried a larger or different such component. It is also possible that the two diverged mtDNA lineages seen in Denisova 8 on the one hand and Denisova 3 and Denisova 4 on the other were both introduced into the Denisovans from this unknown hominin, as has been suggested for the mtDNA of Denisova 3. However, more nuclear DNA sequences from Denisovan specimens of ages similar to Denisova 4 and Denisova 8 are needed to address this question fully.".

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2015 by Austin Whittall © 

Neanderthal in America, data from an interesting genetics paper

A recent paper (G. Povysil and S. Hochreiter, 2014) [1], analyses "very short identity by descent (IBD)" genetic segments of chromosome 1, comparing those in humans, Denisovans and Neanderthals.

Their findings are contradictory and disclose some oddities among "Admixed Americans". By the way, it is a pity that they did not include real "South American Indian", South Asian and Oceanian genomes (the Puerto Rico, Colombian and Mexican populations that were included in their study are not pure Amerindian, they are a "Admixed Americans", a mixture of European, African and Native American peoples - more on this below). Nevertheless the paper is interesting and shows very close links between Africans and Americans that cannot be attributed to post-discovery mixture (slave trade) and some incongruent findings.

IBD Segments

When individuals have the same -identical- nucleotide sequences in a given segment, this segment is said to be identical by state (IBS). When these individuals inherited this segment from a common ancestor, it is said to be identical by descent (IBD). The ancestor of those individuals carried this IBD segment and passed it on to them.

How can you tell apart an IBS from an IBD? That is, two or more persons could be IBS, sharing an identical sequence yet this may have arisen purely by chance and not by inheritance, so they would not be IBD.

The authors looked for "rare variants" which as their name indicates are not very common and therefore if a rare variant is an IBS it is highly probable that it is an IBD too. Which makes sense.

They also found some IBD shared with Denisovans and Neanderthals, so these are also very ancient.

Their findings...

So what did they discover? some interesting facts, below are the highlights:

• Denisovans and Asians. The IBD segments of Asians were the ones with the closest match to the Denisovan ones. Some segments were exclusive to Asians. They were longer than other segments shared by other populations confirming an "In Asia admixture".
• Neanderthals and Asians. The same as with Denisovans: Asians have the highest sharing with Neanderthals and the segments are longest and many are exclusively Asian. The Europeans also share a high frequency of Neanderthal segments but... theirs are also found in other populations too.
• Africans. Unexpectedly, they have many Neanderthal and Denisovan IBD segments, and some are exclusive to Africans. These must be ancient, and originating in the common African ancestors of H. sapiens, Neanderthals and Denisovans.

Let's dig deeper into these generalisations:

On Pearson Correlations

First, some theory. To measure the strength of a linear association between two variables, a statistical correlation is used: the "Pearson product-moment correlation coefficient" (symbolized by "r"). What it does is draw a line of best fit through the data of two variables and "r" indicates how far away from this line are these data points.

"r" can range from -1 to +1. The closer to 0, the worse the correlation, closer to 1 or -1 the better the correlation.

For negative values, the association is negative (one variable grows, the other drops), and for positive values both grow.

Below are some correlations as an example. The case where r=0, it is clear that there is no correlation between variables. (i.e. "eye color" and "criminality" would have a similar correlation: none). However "poverty" or "unemployment" and "criminality" will have 0 < r<1 (there IS a strong correlation between those variables).

Below are some examples of Pearson correlations (I prepared them with my Excel sheet):

Some examples of Pearson correlation. Copyright © 2014 by Austin Whittall

Having said this, let's see the Pearson correlation coefficient found for IBD and populations:

Denisovans

Figure 1, Adapted from [1]. Pearson correlation between
populations and the Denisova genome

ASW: AFR Americans with African ancestry from SW US, YRI: AFR Yoruba in Ibadan, Nigeria, LWK: AFR Luhya from Webuye, Kenya, CLM AMR Colombians in Medellin, Colombia, MXL: AMR Mexicans from Los Angeles, California, PUR: AMR Puerto Ricans, CEU: EUR Utah residents with ancestry from northern and western Europe, FIN: EUR Finnish, GBR: EUR British from England and Scotland, IBS: EUR Iberians from Spain, TSI: EUR Toscani in Italy, CHB: ASN Han Chinese from Beijing, CHS: ASN Han Chinese from South, JPT: ASN Japanese from Tokyo, Japan

The figure above shows four regions with different population groups (Africa, America, Europe and Asia). The Africans show a negative correlation, Asians, Europeans and Americans a Positive one.

The correlations are not all that "strong": Asians average: 0.25, Europeans: 0.1, "Admixed Americans": 0.025 and Africans: -0.075. Which in my opinion are quite weak (very close to zero).

The Asians have the "highest correlation" to the Denisovan genome. Among Admixed Americans, the Mexicans (MXL) have a high correlation too; the authors point out that: "Mexicans (MXL) have also a surprisingly high correlation to the Denisova genome while Iberians (IBS) have a low correlation compared to other Europeans" [1]

This MXL - IBS anomaly is explained as follows: "of all European populations Iberians show the highest rates of African gene flow whereas Mexicans show a high proportion of Native American ancestry which in turn might also reflect gene flow from Asia..." [1].

Allow me to disagree (see my argument further down).

They believe their findings corroborate other studies where Denisovans share more genes with "modern East Asians and South Americans (called Admixed Americans here) than with Europeans". [1]

Now, let's see what they have to say about Neanderthals....

Neanderthals

Figure 2, Adapted from [1]. Pearson correlation between
populations and the Neanderthal genome

Again we see the same trends as with Denisovans but notice that the correlations are even stronger (the values are higher in absolute value but still ‹ 0,5). Again too, the Mexicans have a better correlation than Iberians, which is (again) lower than the rest of the Europeans.

Asians average: 0.4, Europeans: 0.2, "Admixed Americans": 0.1 and Africans: -0.15. Less weaker than the Denisovan correlations.

According to the authors, "As expected, Asians again show the highest odds for IBD segments matching the Neandertal genome... while Africans have the lowest odds .... Europeans show clearly more matching with the Neandertal genome than Admixed Americans." [1].

Why should Admixed Americans have less than Europeans? They are a mix of roughly 50/50 Europeans and Asian (via Beringia) - the touch of African (via slave-trade) is negligible as we will see further down- so Admixed Americans should have IBDs somewere in between those of Europeans and those of Asians... but they don't.

The authors [1] attribute the North-South decreasing cline in Europe (with lowest values in Italy and IBS -in Spain) to "reduced rates of shared ancestry compared to the rest of Europe... [and]... higher IBD sharing between North Africans and individuals from Southern Europe which would decrease the amount of DNA sharing with Neandertals." [1]

In other words, Africans (with less Neanderthal genes) mixed more intensely with southern Europeans thus watering down any Neanderthal IBDs in their genomes (even though Spain and Italy were peopled by Neanderthals and a place where admixture with modern humans should have taken place....

The Odd "figure 10"

Figure 10, from [1]

Figure 10, shown above, perplexes me. The caption in the paper corresponding to this figure is the following (and I highlight the "baffling" text):

"Figure 10: For each genome and each IBD segment, the color indicates whether a population contains this segment (“With”) or not (“Without”). For the human genome, 4,000 random IBD segments were chosen. IBD segments that match the Neandertal or the Archaic genome are found more often in Asians and Europeans than all IBD segments (human). This effect is not as prominent for IBD segments that match the Denisova genome." [1]

In other words, they checked the "content" of specific genomes (the human one, the Neanderthal one, the Denisovan one and the Archaic one) in different populations and showed if the populations carried them or did not carry them in their current genomes.

In the text they state: "IBD segments that match the Neandertal or the Archaic genome are found more often in Asians and Europeans than all IBD segments (human genome). This effect is not as prominent for IBD segments that match the Denisova genome, but still significant." [1]

What I think they are trying to say is the following (take a look at fig. 10 above please):

Asians and Europeans have roughly 200 - 300 IBD segments that are "human", 1,500 that are Neanderthal, some 400 that are Denisovan and about 800 that are Archaic. The "old" genome (2,700 segments) is larger than the "human" one (200- 300 segments).

What they don't point out is that Americans (orange bar in fig. 10) have much higher numbers of ALL segments than Europeans or Asians, and in the case of Neanderthal IBDs, they have the highest values of all populations, including Africans. Africans also surpass both Europeans and Asians... so what is so striking about the Eurasians anyway? What is striking is the high content of "ancient" IBDs in Admixed Americans and Africans.

Could this mean for instance a very ancient peopling of America an Out of Africa directly to America (ie. H. erectus), or later, with Neanderthals, long before modern Humans reached the New World?

Americans also have a much higer "human" IBD segment content than Europeans and Asians, second only to Africans

Note that the “Archaic” element is much more significant than the Denisovan one in both Europeans and Asians... is this due to archaic genome carried by Neanderthals or the Sima de los Huesos ancestor of Denisovans? or, what about the Dmanisi in Georgia or even Homo erectus in Southern Eurasia?

Noticing the closeness between African and Admixed Americans, the authors quickly point out the "African" admixture in Americans (slave trading to the Americas after its discovery). This is, in my opinion, an incorrect assessment of the situation.

Mexican and Iberian anomalies, the correct interpretation

First of all, a "MXL" or "Mexican from Los Angeles, California" is a rather untidy way of studying traces of ancestral hominids in America... it is like looking at say, "Italians in Buenos Aires" or "Dutch in Cape Town" or even "Ashkenazi in New York", none of these groups, real and relevant as they may be, reflect the original people of those areas (Querandí, Khoi or the Lenape, respectively).

MXL, and the other "Americans" (CLM, Colombians in Medellin, Colombia and PUR Puerto Ricans) are an admixture of European, African and Amerindian. So it is difficult to come to any conclusions. Europeans, Africans and Asians on the other hand, are the "original" peoples of their respective continents and can tell us about the changes in the genomes of their respective geographic areas.

Nevertheless, we will try to work with these artificial constructs, the MXL, PUR and CLMs:

I came across a paper that takes the 1000 Genomes project data (Gravel et al., 2013) [2] and gives us a breakdown of the makeup of those "Admixed Americans" which is shown below: [2]

• MXL. 47.6% Native American,  4.2% African, 48,2% European.
• PUR. 12.8% Native American, 11.7% African, 75.5% European.
• COL. 25.6% Native American,  7.5% African, 66.9% European.

Being people originating in former Spanish Colonies (Colombia, Puerto Rico and California were part of the Spanish Empire until the XIXth century), the European content is predominantly Iberian, and mostly Spanish (IBS), which may also have some ancient admixture with African individuals in their genome [3].

We must ask ourselves why the correlation with Neanderthals and Denisovans in Admixed Americans is similar to the correlation of IBS when only half their genome is Iberian? It should differ because the other half is predominantly Amerindian, allegedly closer to Asians than to Africans. Since Asians are even more correlated to Neanderthals and Denisovans we must ask ourselves how come the correlation of Admixed Americans isn't higher than the European one?

I used a rough approximation to calculate the value for the admixed Americans (very rough and I am inclined to believe that it is flawed): I calculated the weighed average by multiplying the "Pearson correlation" of each of the populations that originated the different admixed Americans by its weight in each Admixed American group, and adding them up to obtain a "calculated Pearson" for both Denisovans and Neanderthals for that group:

Example Neanderthal Pearson among MXL. See the image below (in brackets I indicate each of the populations that make up the MXL):

0,48 x 0.2 [EUR] + 0.04 x -0.15 [AFR] + 0.48 * 0.4 [NAm]= 0.28

For Nam (Native Americans or Amerindians) I took the same value as their alleged ancestors: Asians.

Calculation of Pearson correlation for Admixed Americans
Copyright © 2014 by Austin Whittall

In red, under "Average" is the value for both Neanderthal and Denisovans (0.23 and 0.129 respectively) the average is simple since the size of each American group is nearly the same (see Appendix B in [1]).

When compared with the "Actual" values given by G. Povysil and S. Hochreiter [1], we find that the Denisovan figure is very similar but the Neanderthal value is definitively off mark. Yes, I know that the calculation is rough and indeed surely wrong (see Appendix I at the bottom of this post to see why).

But I disagreed, even without any calculations, because a population that is basically a mixture of Asians (via Beringia) and Europeans must have a Pearson correlation somewhere between the values of both of those two populations. But, the value given by Povysil and Hochreiter is ten times lower! and almost zero (no-correlation)

Furthermore, the data in figure 10 clearly shows that Americans have the highest Neanderthal IBS of all groups.

There is something that is not quite right in this paper or I have not understood the data.

Additionally, Table 1 in [1] shows how IBDs are shared between populations. And it has some striking values:

Key: AFR = African, AMR = Admixed Americans, EUR = Europeans, ASN = Asian.

• AFR / AMR = 28,710
• AFR / ASN = 387
• AFR / EUR = 986
• AMR / ASN= 207
• AMR / EUR =1,008
• ASN / EUR = 351

Once again, as with Fig. 10, Africans and Americans are close together: Africans and Americans share nearly 29,000 IBS despite AFR being only 4 -12% of AMR ancestry, while Europeans (48 - 76% of AMR ancestry) only share 1,008 IBD with Americans, and Asians, share even less IBD (207) than EUR.

Africans share 74 times more IBS with Americans than with Asians and 29 times more than with Europeans. Why? Slave trade cannot explain this. Maybe an ancestral admixing (H. erectus or even earlier... Dmanisi?)

In Appendix A [1] , they also indicate the "rare and low-frequency variants .... (0.5 - 5% frequency)", found in different quantities in each population. As expected (by me) Africans and Americans have the highest values:

• AFR: 558,996 - 683,289
• EUR: 143,987 - 155,270
• ASN: 118,137 - 120,115
• AMR: 163,737 - 214,430

The authors note that Africans have 4 times more rare alleles than all other populations and add, trying to attribute America's high values to admixed African genes, they state (bold mine): "if we ignore the Admixed Americans that have African admixture".

But this statment is off mark: we have seen that Admixed Americans only have 4 - 12% African genetic content!, how can they have such a high frequency of rare alleles? They have +28% more than Europeans and +60% more than Asians! These are supposedly "rare" and "ancient" markers, shouldn't they appear more frequently in the Old World?

I even suspect that the massive loss of lineages in America post-discovery due to disease and war during the XVI century, wiped out many rare lineages and what we see is the tip of a once vast iceberg. The ghost of the ancient peopling of America.

It is a shame that the paper did not include "real" Amerindian and Papuan / Oceanian samples. Maybe future studies will do so. It is also a pity that when faced with a brink, they back off and stick to orthodoxy.

Appendix I

Why my "rough calculation" was flawed:

Actually Admixed Americans are a mixture of "certain" members of three different populations, whose admixture can range from 0 to 100% and include any of, or all of, these populations. That is, for example: "pure" american natives with no European or African ancestry on one hand and, mixed descendants of Europeans and Africans and American Natives on the other.

So the calculation is not a simple mathematical average. The resulting distribution depends on which members of the orignal populations take part in the final admixture. Because the individuals of each population carry different frequency of Neanderthal or Denisovan alleles in their genomes which may be different to that of the whole group.

Just as an example I made a simulation with three populations (American, African and European), each with a distribution that gave a Pearson correlation (r) identical to those reported by Povysil & Hochreiter [1] for Neanderthals, the outcome is shown below:

A simulation of Correlations by admixing populations, for Neanderthals and MXL
Copyright © 2014 by Austin Whittall

On the left side I placed three clusters: the Europeans are shown with blue rhombus, their trend is the blue line with r=0.2; the Africans are the red squares and the red line marks their r=-0.15, finally American Natives are the green triangles with a green line marking r=0,4 (For them I adopted the Asian values).

Note the dispersal of the dots (a cloud of dots with no real pattern), this shows how "slack" correlations are when they are far from 1 and close to 0.

On the right side is a mixture, which takes the Mexican "combination" of 48% European, 4% African and 48% Amerindian. So if there were 16 values in each population, the mix was made up of 8 chosen from Europeans, 8 from Amerindians and 1 from Africans. These points were taken from the original distributions shown on the Left, and the outcome is shown with the red circles and the red line. I chose any points, at random.

The Pearson correlation value in the chart, for the mixed population, was r=0,11. But in other simulations, depending on the points chosen from each population, I obtained different correlations which varied widely, from r= -0,08 to r= +0,83.

This proves that my assumption was flawed. Nevertheless, the populations used by [1] consisted of more than 16 values or points, they had over 350 individuals in each population. So the dispersion "cloud" would not be similar to mine and it is likely that the final values of "r" would not be so different.

Sources

[1] Gundula Povysil, Sepp Hochreiter, (2014). Sharing of Very Short IBD Segments between Humans, Neandertals, and Denisovans. bioRxiv April 7, 2014. doi: 10.1101/003988
[2] Simon Gravel et al., (2013) Reconstructing Native American Migrations from Whole-Genome and Whole-Exome Data. PLoS Genet. Dec 2013; 9(12): e1004023. doi: 10.1371/journal.pgen.1004023
[3] Johnson NA, et al., (2011). Ancestral Components of Admixed Genomes in a Mexican Cohort. PLoS Genet 7(12): e1002410. doi:10.1371/journal.pgen.1002410

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2014 by Austin Whittall ©