San (Khoisan) and Europeans

Following my previous post on similarities between San people in southern South Africa, and Europeans, I read a paper about the San (Khoisan) and their "antiquity". It reports them as an ancient population that retained a large effective population while that of other groups fell (i.e. other Africans, Europeans, Asians, and the Out of Africa migrants).

The paper published in Nature in 2014 by Kim et al., (Khoisan hunter-gatherers have been the largest population throughout most of modern-human demographic history). The image below shows how Ne (effective population) evolves over time (oldest to the right), for San, African Yoruba, Europeans, and Asians. As you can see all groups (actually, the ancestors leading to each of these populations) have similar population sizes till 100,000 years ago when a dramatic drop in population sizes occurs. I don't understand how they obtained Ne values for hominin populations 2 to 4 million years ago, this was the days of Australopiths and probably Homo habilis.

Effective population sizes for San, Africans, Asians and Europeans. Fig. 3 a in Kim et al., 2014

This drop in effective population size is attributed to climate changes within Africa. The paper includes a series of maps as Fig. 12 in its Supplementary Material to explain the process. They can be seen below:

Modern humans originated in Africa (a), blue circle in South Africa seems to imply an origin there, though the paper does not specify the location. Then these people spread north (b), the orange oval marks the new territories. Then came the climate change (c) around 150 or 100 kya. Drought in western and central Africa hit the humans there in central, western, and eastern Africa, but spared the San people in the southern part of Africa. This coincided with a fragmented population (structured) with isolated groups that did not interact with each other (see the different dots and colors on the map, marking these groups). Populations declined central and western Africa, and when the ancestors of Non-Africans (green arrows) (d) migrated Out of Africa (OOA), they carried this lower Ne, and it dwindled even more due to bottlenecks and founder effects as they advanced into Eurasia. The San, however, kept their population intact.

The authors reconstruction of this period is summarized as follows: "After the earliest split, between the ancestral Khoisan and non-Khoisan populations ~100–150 kyr ago, the ancestral Khoisan population maintained their high genetic diversity, while the effective population size of the non-Khoisan continued to decline for 30~120 kyr ago and lost more than half of its diversity. The ‘Out of Africa’ migration ~40–60 kyr ago accounts for the observed population split between African and non-African populations, and the subsequent smaller effective population size of non-Africans compared with non-Khoisan Africans."

Comments

However, and interestingly, as pointed out in my previous post, the San and Europeans share several unique allele variants that are ancestral (found also in Neanderthals and Denisovans) whcih confer lighter pigmented skin than that found among the remaining Africans and also South Asians and Australo-Melanesians, who carry a later (derived) mutation for darker pigmentation.

How does this similarity between a specific OOA group and San people tie in with the evolution and migration sproposed by Kim et al.?

Not well. We would have to imagine a group that split from the San, moved north, lived in isolation in Central Africa, then survived the climate crisis there, moved north, left Africa, surviving the founder effect, bottlenecks and genetic drift, established themselves in Europe and somehow managed to keep their skin-color alleles intact. While all the other groups in Africa mutated and adopted a dark skin set of alleles. Too complex to be the explanation.

The San (Khoisan or bushmen) have always intrigued me since the 1980s movie "The Gods Must Be Crazy", I was taken aback by their pale skin and oriental factions. So different from the usual African features. People living in the deserts of Namibia with a hunter-gatherer culture in the 20th century! I have never found a paper explaining their similarity with East Asians. I will explore this strange trait in a coming post.

---

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

High African diversity despite population bottlenecks. Why? (archaic admixture in Africa)

Higher diversity is normally associated with a effective populaton size, and that is why the Out of Africa migrants are said to have a lower diversity: they moved in a small group and this carried less diversity than the original population (a founder effect and a bottleneck for the migrants),

However, a paper published in Nature by Jakobsson et al., 2025 studied the evolution of modern humans using genetic information garnered from "ancient southern African genomes" and found that the African population size wasn't all that big. In fact it was the same size as the Out of Africa band!

Below I quote the relevant passage in the paper, my comments in brackets, and I highlighted some interesting parts of the text:

"Long-term large population size
... Heterozygosity (HO) for ancient southern Africans (mean across genomes; HO = 0.80 × 10−3) was similar to other ancient Africans, only surpassed by an ancient western African individual (HO = 0.93 × 10−3), indicating a large Holocene population size in southern Africa. A multiple sequentially coalescent approach shows that the effective population size (Ne) was large for several hundred thousand years, up to Ne ≈ 30,000 around 200 ka, similar to other African groups. The large Ne at ≥300 ka for all humans was potentially caused by population subdivision. [interesting! so small populations divided into many give the appearance of a large Ne when in fact it isn't] We note a decline in Ne for ancient southern Africans from around 100–50 ka, to Ne ≈ 10,000 by the Last Glacial Maximum (20 ka), similar to non-African groups and the ancient northern Africans [So at the time of the supposed Out of Africa Event 100,000 to 50,000 years ago southern & northern Africans and non-Africans had the same population size! So where is the OOA bottleneck?]
Runs of homozygosity (ROH, where greater numbers and total length of ROH segments indicate a smaller population size) show that the ancient southern Africans were at the upper tail of the distribution of modern-day Africans, but less extreme than most non-Africans—a pattern attributed to the out-of-Africa bottleneck. [homozygosity is attributed to small populations, and inbreeding, loss of diversity or heterozygosity. So the supposed most ancient humans, the South African San people have the highest ROH among Africans — but lower than non-Africans.] This indicates a smaller population size (relative to, for example, western African groups) in the relatively recent history of each individual, but still larger compared with non-Africans and ancient northern Africans. Most ancient southern Africans are shifted towards greater total segment ROH length without affecting the total number of ROH segments, in particular the Great Brak River (2,355–2,310 cal. bp) and the Matjes River 1 (7,845–7,690 cal. bp) individuals [this date is extremely recent! well after the OOA event]. This pattern indicates a smaller recent ancestral population size, possibly with elements of inbreeding, indicating isolation and fragmentation among ancient southern Africans during the Holocene. Ancient southern Africans south of the Limpopo River therefore consisted of a large, stable population for many millennia, with a modest decline since around 50 ka, and a possible fragmentation and further decline during the Holocene."

Admixture with archaics!

As usual the paper also points out that: "Population stratification between southern Africa (the region south of the Zambezi River) and the rest of Africa probably existed for at least 300 thousand years (kyr), perhaps up to a million years. Such deep stratification may result from admixture with an unknown archaic African group predating the divergence of Homo sapiens from Neandertals and Denisovans, and//or from isolation from other groups." The isolation would mean that an archaic form of humans didn't admix with the others and carried ancient, unshared alleles. The admixture option seems more plausible.

When discussing diversity, the authors note that: "Cumulatively, the genomes of the ancient southern Africans show that this group displays many Homo sapiens-specific variants (and variable positions) at amino acid-altering sites, also reflected among the modern-day San people. This observation cannot be explained solely by a large, stable southern African population, which retained derived variants to a greater extent compared with other groups. The ancient southern Africans were probably also isolated from other African groups for long periods. The derived variants unique to southern Africans may also signal low-to-modest gene flow from an unknown/unsampled group of genetically differentiated humans." Indeed, introgression from archaics.

Notice how reluctant mainstream scholars are. Instead of digging deeper into the archaic admixture hypothesis, they set it aside. In this case, the authors seem to agree with this option, but, politely wrote: "Irrespective of cause, the many variable amino acid-altering sites among the ancient southern Africans point towards a genetic model in which different protein variants can be combined to viable outcomes... The many Homo sapiens-specific variants found in southern African genomes point to a combinatorial genetic model of human evolution in which there are many possible combinations of genetic variants that lead to ‘genetically modern’ Homo sapiens." Yes, a combination brought about by mating with archaics within Africa.

---

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

Another paper on Introgressions (April 2026)

Continuing with the wide variety of introgression / admixture papers published over the past few years, today I add a new preprint (in Biorxiv, and therefore not peer-reviewed) published a few days ago, on April 12, 2026: Inferring hominin history with recurrent gene flow from single unphased genomes and a two-locus statistic. Nicholas W Collier, Simon Gravel, Aaron P Ragsdale. bioRxiv 2026.04.11.717825; doi: https://doi.org/10.64898/2026.04.11.717825

Through the use of a very particular statistical model (described at the beginning of the paper, and well over my statistical abilities to understand), and genetic analysis of the autosomal DNA, the authors suggest a population structure and admixture, and population sizes for modern humans, Neanderthals, Denisovans, and super-archaics that mix to and fro over the past million years. The paper assumes "a fixed mutation rate of 1.3 × 10−8 per bp per generation and a generation time of 29 years" (I have previously posted about mutation rate, its variability, and generation times and the combined effect of them on calculating timelines.

The paper reports the following events and dates:

• Neandertal-Denisovan Common ancestor or ND lived from 779 to 726 kya and lasted for ~50.000 years.
• Ancestral Neanderthals or AN that around 123 kya split into Altai people in Siberia - who later became extinct, and the Western Neandrthals or WN
• Anatomically Modern Humans or AMH introgressed into AN 250 kya ago, and 110 kya into the western Neanderthal (WN) group, which later evolved into the Croatian Vindija and Chagyrskaya (Siberia) lineages.
• Denisovans received gene flow from a ghost lineage, a "Superarchaic" S that may be Homo erectus, it had split from our ancestors 2 million years ago.
• They reckon that the Ust’Ishim people from East Central Siberia dated to around 45 ky were the first humans in Eurasia to split from the other branches after the Out of Africa Event.

The arrows in the chart show the introgression: "broken one-headed arrows denote instantaneous gene flow events; solid double-headed arrows denote continuous gene flow." The percentages, and population sizes (N_e) are also represented:

Figure 6: Early hominin history in Eurasia with recurrent gene flow. From Nicholas W Collier, Simon Gravel, Aaron P Ragsdale, 2026.

The timeline is the following:

TND→AMH (ky) AMH–ND split time 798 CI: 748 – 827
TAN→Den (ky) AN–Denisova split time 688 CI: 639 – 734
TWN→Alt (ky) WN–Altai split time 123 CI: 117 – 137
TCha→Vin (ky) Chagyrskaya–Vindija split time 60.5 CI: 57.3 – 67.7
TYor→OOA (ky) Yoruba–OOA split time 56.9 CI: 53.6 – 60
TOOA→BE (ky) OOA–BE split time 54.7 CI: 49.7 – 57.6
TLos→Stu (ky) Loschbour→Stuttgart admixture time 29.4 CI: 13.2 – 35.9

The authors conclude that "Using these advances, we inferred a demographic model that broadly explained observed H2 patterns and integrated major supported features in hominin evolution, including recurrent interbreeding between Neanderthals and AMH, introgression from a distantly-related, unsampled lineage to Denisovans, and population structure in western Eurasian AMH."

There is no Denisovan to AMH admixture in this model, it seems to only focus on Northern and Western Eurasians, and does not consider Eastern, Southern or Southeastern Asians and Oceanians.

Effective Populations

I found the Effective population sizes to be of interest (the N_e). As you can see, the Ancient basal root at the top of the image (A) has a large population from which the Superarchaics (S), the modern humans (AMH) split from and conserve a large population size, the ancestor of Neanderthals and Denisovans (ND) has a tiny population and remain that size, so do the N, Denisovans, and the original Out of Africa migration group (bottleneck). The Yoruba people retain a large population.

The paper says that the original Ancestral population had an N_e of 16500 individuals (CI: 15900 – 17300) and then it says "We fixed the effective size of the Superarchaic lineage (S) to 20,000" So the superarchaics splitting from the Ancestral line into Eurasia didn't suffer a bottleneck? Why?

Yet the other groups splitting from the Ancestral group did! The authors explain this large Superarchaic population size as follows: "We justified fixing the population size of S with the observation that changing the effective size of a ghost lineage which makes a small ancestry contribution to a sampled lineage has a negligible effect on E[H2]." So, their model and formulation allows these unrealistic assumptions.

The upper part of the image further down, shows how the Superarchaic introgression into Denisovans affects the population sizes and dates, their model calculates an outcome with a minor impact on effective populations or the timelines.

Interestingly, they note that small effective population sizes may be an artifact, because they can be "plausibly explained by geographic population structure. With spatial structure, recent ancestors are expected to live in closer proximity, and to therefore have a higher probability of sharing parents, than ancient ancestors. Strong structure therefore causes recent coalescence rates to be larger than ancient ones,a pattern which is interpreted as a small recent effective size in a panmictic model."

The paper also notes that "using a lower mutation rate inflated effective size and time parameters, while a higher rate diminished them." They show tables with the effects of different mutation rates as can be seen in the lower part of the image below. The image shows the effects of Superarchaic introgression into Denisovans and the effect of different mutation rates on Ages and N_e of the hominin clades. The original can be seen in tables S8 and S9 in the Supplementary Information of this paper:

The impact of a slower mutation rate can be seen in the older split between ND and our lineage and an older ND split into Denisovans and Neanderthals, but does not affect on more recent events. The impact of mutation rates on the effective population size (N_e) is also variable, some populations have a bigger effective population (A, AMH, Altai, Vindija, Chagyrskaya, Denisovan) while others smaller (ND, Yoruba).

I have already posted about mutation rates and how it impacts on N_e and heterozygosity, and mentioned the same effect reported by the authors of this paper: For a given heterozygosity, lower mutation rates increase effective population size. I also posted about the effect of mutation rates on dating splits, lower rates lead to deeper (older) split dates.

However, this paper does not explain why the same mutation rate affects N_e in opposite ways (there is a complex explanation about their model in the Appendix that mentions some effects on the Effective Population size). It does admit that the model, like all models is a simplification of reality: "... we made many approximations to simplify our models. We treated populations as discrete entities, with random mating, piecewise-constant sizes, and instantaneous divergence. Some of these assumptions allow us to model the evolution of HR statistics, while others are useful for formally testing tractable demographic models. Of course, the true evolutionary history includes unmodeled populations, continuously fluctuating population sizes, population structure induced by the spatial distribution of individuals, and variable migration rates... We also made a number of simplifying biological assumptions. We assumed that the genome-wide average germline mutation rate was constant across all lineages throughout the modeled period... We also assumed a constant generation time for all lineages throughout the studied period."

Interesting work.

Introgression Index

Visit my index post, with all the introgression posts in one single place.

---

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

Neutral Theory of Genetic Evolution and Out Of Africa

The main backing for the Out of Africa theory is the Genetic Neutrality Theory.

The arguments of an African origin of modern humans and our dispersal across the globe is supported by the high genetic diversity found in modern African populations, with lower diversity elsewhere, and a gradient or cline in diversity that reflects less diversity as distance to the African homeland increases. Both of these factors are expected according to the Neutrality Theory.

Starting with a highly diverse population, if a small group from that population migrates (into Eurasia), it can only feasibly carry with it a sub-sample of the original diversity. This is known as a Founder Effect, the founders of a new population carry fewer genes than the population from which they split from.

This happened time and time again, as sub-sub-groups split from the main population and moved into Europe, Eastern, Northern, and Southern Asia, Melanesia, Australia, Polynesia, and across North America, and then, into South America.

The Neutral Theory states that each split reduces genetic diversity.

Genetic Heterozygosity

Heterozygosity is a measure of diversity. Each person receives genes from their parents, that code for different proteins and produce traits. Those who have two different varriants (alleles) of a specific gene, one inherited from each parent are heterozygous. If the alleles are identical, they are homozygous.

The image below shows two parents (both are heterozygous) each carrying two different variants A and a. The probability for passing them on to the next generation is simple there are four possible combinations, each has a 25% probability of occurring:

Hetero and Homozygosity. Copyright © 2026 by Austin Whittall

The chances are that two of the offspring will carry Aa alleles, and will therefore be heterozygous, while the other two will receive the same allele from each parent and be either AA or aa, carrying two identical copies. This makes them homozygous.

As we can see, a population that is 100% heterozygous as become 50% homozygous and 50% heterozygous. All the possible combinations of those homozygous and heterozygous genes are shown below:

Combinations of alleles. Copyright © 2026 by Austin Whittall

As you can see 25% of each variant (AA, aa, Aa, and aA). So why would heterozygosity decrease? Suppose only aa homozygous couples mate, the chance of this happening is 1 in 16, or AA mate, again, 1 in 16. So 2:16 or, 1:8 chance of only homozygous mating and offspring. But... if these offspring meet and mate aa with AA, they would have a 100% heterozygous descent. This is true for large populations, but for smaller groups the founder effects and bottlenecks can reduce the allele diversity.

Genetic Bottlenecks

The argument of loss of heterozygosity, or its equivalent, increase in homozygosity is based on genetic bottlenecks, where a small sub-population splits and carries with it the homozygous variant, say only aa or only AA. Losing the possibility of reintroducing the lost allelle. This is a 1 in 16 chance.

Other causes of heterozygosity loss are natural catastrophes, war, and disease. But, why would such events affect the heterozygous individuals more than the homozygous. Wouldn't they be random, and therefore have an equal chance of impacting on hetero- and homozygous individuals?

Regarding the root population. There is the chance that the root from which a population split off from suffered some event that eliminated a large swath of it, while the migrating sub-population in another geographic location was not affected by it. Wouldn't that lower the heterozygosity of the basal group and make the sub-population appear as "enriched"?

Genetic Drift

Both Founder effect and Bottlenecks are part of process called Genetic Drift. As we saw, genetic drift takes place when random events, by chance modify which alleles passed on by parents to their offspring. They also include not only non-reproduction of certain individuals due to war, disease, natural catastrophes, but also loss of genetic variation due to people who don't reproduce because they die before mating, choose not to do so, etc. Genetic Drift isn't driven by evolution. The random changes may or may nor provide adaptations to a changing environment, so they may or not be acted upon by the forces of natural selection.

A sub-population may lose certain alleles, or others may become Fixed reaching a 100% frequency in the population due to chance events.

Mutations and Natural Selection

Random mutations take place, and modify the alleles, natural selection may also work, favoring the survival of individuals with alleles that provide adaptative benefits.

But, what about mutations, that happen by chance, that have a deleterious effect? Some mutations may have harmful consequences. The Neutral theory says that some deleterious mutations may rise to high frequencies in small populations due to fixation promoted by genetic drift. But, why wouldn't people carrying unfavorable genes be affected by natural selection, causing them and their descent to die out?

The Neutral Theory of Molecular Evolution

It was the creation of Motoo Kimura, who in 1968 proposed that at a molecular level, mutations are caused by random genetic drift. These mutations are neutral from a selective point of view. They aren't affected by natural selection.

Kimura has been criticized, for instance Kern and Hahn (2018), argue that modern, genome-scale data demonstrates far more evidence of adaptive evolution than the neutral theory allows, suggesting that natural selection (both positive and negative) shapes much of the genome.

As mutations take place by chance, the probability of them being neutral, deleterious, or beneficial would seem equivalent. So, why assume they are neutral? A beneficial mutation even if it is rare would confer an evolutionary advantage for those carrying it, and modify the population beyond what neutral models suggest.

Linked Selection. The loci (or addresses) that mark the location (locus) of a gene in our DNA isn't independent and isolated. Some genes or DNA sequences located close together on the same chromosome are inherited together, as a unit, during meiosis (linked chromosomes).

Selective Sweep is when an allele that improves the fitness of its carrier increases in frequency due to natural selection, is accompanied (hitchhiking) by other genes linked to it by physical proximity on the DNA strand are also increased in frequency even though they may be neutral. Finally, Background Selection is similar and has the opposite effect: deleterious alleles are removed by natural selection and neighboring neutral alleles are lost too, due to physical proximity to the harmful variants.

These examples show that "neutrality" is not necessarily true.

Molecular Clock

Kimura's theory states that neutral mutations took place at a constant speed, accumulating over time at the same pace. However, this is not true.

However mutations don't appear in a uniform manner in all loci along the genome, they arise unequally, and the probability of fixation depends on where they arise in the genome. This modifies how the clock ticks (Source). Furthermore, substitutions depend on population size, and generation overlap (Source).

Generation time is also an important factor: is it 20 or 30 years? 25? or 18? Over 10,000 generations this means a time scale that can vary from 180,000 to 300,000 years!

Back-Mutations and Recurrence Not Allowed

Kimura's theory, at least when applied in practice, has three axioms that are not true:

1. Infinite sites, it assumes that each mutation takes place at a site that has never mutated before.
2. No back-mutations, changes happen in one direction A → G. Which will never again flip back G → A
3. No Recurrence, in practice there are multiple mutations that take place at the same site. The neutral theory does not accept it, there can't be multiple mutations at identical loci in different lineages.

A paper gives a great example of why and how a back-mutation can have positive effects (here showing how a base C = Cytosine mutates to T = Thymine and back):

"...simple back-mutation is expected to generate slightly advantageous mutations. For example, let us imagine that a site is fixed for C, and that a new T mutation occurs that is slightly deleterious with a disadvantage of −s. Let us imagine that this T mutation spreads through the population and becomes fixed. If a new C mutation then occurs at this site, it will be slightly advantageous with an advantage of +s, unless the relative fitnesses of the C and T alleles have changed. Such a change in fitness could occur because of a change in the environment or the fixation of mutations at other sites which have epistatic interactions with the alleles at a site of interest."

Americas: Great Dying

Regarding Amerindian diversity, we know that up to 90%, or more, of the Native Americans died during the century that followed European "discovery". Disease, war, famine, social disruption, force labor, etc. killed tens of millions of Amerindians. Lineages died out, massively. This is the unique and most massive genocide (albeit unplanned) in the history of humanity. How can we know the number of unique, diverse, divergent alleles that were wiped out during this event? In 1491, America probably presented a far more diverse genetic structure than it does now.

And this brings us to the other point: African "diversity".

African Diversity... is it real?

Finally, and this will be the subject of a future post, do modern Africans reflect the genetic makeup of ancient Africa 100,000 or 75,000 years ago? Is a modern Nigerian, Gambian, Angolan African representative of the ancient population from which the Out of Africa migrants split? Have other events taken place within Africa, isolated from the sub-population that migrated into Eurasia? Admixture with archaic hominins after the OOA event, admixture between many separate and formerly isolated hunter gatherer sub-populations could have led to a modern highly diverse African population, while the original OOA root was far less diverse.

---

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