Sunday, 9 April 2017

More musings on monogenism (or why the human population was never as small as two people)

If every single human alive was exclusively descended from two people who lived around 6000 years ago, we would not expect the data from population genetics to show that the human population was never smaller than a few thousand individuals. Nor would we expect to see coalescent times for human genes in the hundreds of thousands of years. This is however exactly what we see, and it completely falsifies the anti-evolutionist position that the entire human race descended from six individuals approximately 4500 years ago and propagated away from Mt Ararat. What we see continues to show an African origin for the human race. As always, the problem here is simply one of choosing from an avalanche of papers that show this fact.

Minimum human population has been never smaller than a few thousand

            In 2006, Liu et al noted that both fossil and genetic data supported “a recent single origin of modern humans in East Africa”[1] but noted that apart from this basic fact, there were difficulties in clarifying details of this. They analysed 52 human populations, and stated their data “suggest an initial expansion of modern humans 56,000 years ago from a small founding population of 1,000 effective individuals.”
            Five years later, Li and Durban, using whole genome sequence data from Europeans, African, and East Asian individuals, stated that the genomic evidence showed a marked bottleneck in European and Asian populations between 20-40 thousand years ago of around 1200, and a less marked bottleneck in African populations around 50,000 years ago of around 5700.[2] This pattern of a less marked bottleneck in African populations as compared with non-African populations is reflected in other studies. It is also worth noting that these bottleneck sizes are in the thousands, not 2-6 individuals.

Population size estimate from autosomal (a) and X chromosome data (b). Source: Nature (2011) 475:493-496

The 1000 Genomes Project Consortium published in 2015 their report[3] announcing the completion of their project in which they reconstructed the genomes of 2504 people from 26 populations (Africa, East Asia, Europe, South Asia, and the Americas). Some of the salient points:

·      Individuals from African ancestry populations harbour the greatest numbers of variant sites, as predicted by the out-of-Africa model of human origins
·      The human race has a shared demographic history for all humans beyond ~150,000 to 200,000 years ago
·      European, Asian, and American populations experienced a bottleneck ~15,000 to 20,000 years ago, while the African population experienced a bottleneck of lower severity at the same time of around 4250 individuals

Median PSMC curve for each population. Source: Nature (2015) 526:68-74

Stephan Schiffels and Richard Durban published a paper in Nature Genetics reporting on a new method which allows geneticists to examine genetic history more recent than around 20,000 – 30,000 years ago. In applying their technique to genomic data from nine populations, Shiffels and Durban found evidence for separation of the ancestors of non-Africans from African Yoruban ancestors well before 50,000 years ago. Their method also showed a non-African bottleneck (again, not 6 or 2 people!) and a less marked bottleneck in the African population, just as the 1000 Genomes Project Consortium showed:
We find that all non-African populations that we analysed show a remarkably similar history of population decline from 200kya until about 50kya, consistent with a single non-African ancestral population that underwent a bottleneck at the time of the exodus from Africa around 40-60kya. The prior separation of non-African and African ancestral population size estimates begins much earlier at 150-200kya, clearly preceding this bottleneck, as already observed using PSMC. We will quantify this further below by directly estimating the relative cross coalescence rate over time. In contrast, we see only a mild bottleneck in the African population histories with an extended period of relatively constant population size more recently than 100kya. Between 30kya and 10kya we see similar expansions in population size for the CEU, TSI, GIH, and CHB populations.[4]

Source: Nature Genetics (2014) 46:919-925

The Simons Genome Diversity Project[5] has analysed 300 genomes from 142 populations, and likewise has shown that the human population has never been as small as 2-6 people. Specifically, the authors noted:

  • The population (not individual) ancestral to all modern humans began developing substructure at least 200,000 years ago, with substantive separation of non-African and southern African populations around 131,000 years ago, and substantive separation of non-African and central African populations around 112,000 years ago
  • Non-African population sub-structure dates around 50,000 years ago, consistent with the archaeological evidence for modern human dispersal into Europe and Asia around this time.
  • Minimum human population sizes never dropped below a few thousand people.

Source:  Nature (2016) 538:201-206

Mallick et al caution that the date estimates do not take into account uncertainty about the human mutation rate on which their work (and that of the others cited earlier) depends; this could by 30% higher or lower than the value used. However, this does not allow YECs a way out of this data for two reasons. The first is that increased mutation rates result in data that conflict with known archaeological data. As Schiffels and Durban note:
Our results are scaled to real times using a mutation rate of 1.25×10−8 per nucleotide per generation, as proposed recently and supported by several direct mutation studies. Using a value of 2.5×10−8 as was common previously would halve the times. This would bring the midpoint of the out-of-Africa separation to an uncomfortably recent 30-40kya, but more concerningly it would bring the separation of Native American ancestors (MXL) from East-Asian populations to 5-10kya, inconsistent with the paleontological record.[6] (Emphasis mine)
The second reason is that there exists an alternative method utilising linkage disequilibrium, which allows us to calculate minimum effective population sizes that is independent of the mutation rate.[7] In 2007, Tenesa et al published the first genome-wide estimate of the effective population size by using linkage disequilibrium methods, and arrived at effective population sizes of ~3100 from Eurasian data and ~7500 from African, “consistent with the out-of-Africa theory of ancestral human population expansion and concurrent bottlenecks.”[8]

Source: Genome Res. (2007) 17: 520-526

That this evidence is robust, and overwhelmingly argues against monogenism is not seriously doubted outside of a tiny fringe of Christian fundamentalists. As the geneticist Dennis Venema notes, in a 2010 paper in Perspectives on Science and Christian Faith:
Studies based on SNP/LD approaches have now estimated ancestral population dynamics for various human groups over time in more detail than is possible with mutation-based estimates. African groups have a higher effective population size (~7,000) than do non-African groups (~3,000) over the last 200,000 years. This approach, though based on methods and assumptions independent of previous work, nonetheless continues to support the conclusion that humans, as a species, are descended from an ancestral population of at least several thousand individuals. More importantly, the scalability of this approach reveals that there was no significant change in human population size at the time modern humans appeared in the fossil record (~200,000 years ago), or at the time of significant cultural and religious development at ~50,000 years ago.
Taken individually and collectively, population genomics studies strongly suggest that our lineage has not experienced an extreme population bottleneck in the last nine million years or more (and thus not in any hominid, nor even an australopithecine species), and that any bottlenecks our lineage did experience were a reduction only to a population of several thousand breeding individuals. As such, the hypothesis that humans are genetically derived from a single ancestral pair in the recent past has no support from a genomics perspective, and, indeed, is counter to a large body of evidence.[9] (Emphasis mine)

Long coalescent times for genes: more evidence against monogenism

            The coalescent time for a particular collection of variant forms of a particular gene is the time (known as Tmrca in the literature) that has passed since the existence of the most recent common ancestor for that collection of genes.[10]  Needless to say, we would not expect to see Tmrca for genes older than around 6000 years if we were descended exclusively from two people who lived around 6000 years ago.
What are the coalescent times for human genes? Space precludes a complete listing given that there are approximately 20,000 protein-coding genes, but what the literature shows is that the Tmrca for many genes and genetic elements are far older than 6000 years; we are talking hundreds of thousands of years, and in some cases well over one million years. (Given that the oldest Homo sapiens fossil is around 200,000 years old, these Tmrca data alone are enough to show that our ancestry stretches well into the remote past). Laurent Excoffier[11] summarises a number of sequence studies, and shows the Tmrca values for them. Values range from 39,000 years to around 1.78 million years:

It hardly needs to be pointed out that no one is interpreting this data as supporting universal recent human descent from six people who lived 4500 years ago, dispersing from Mt Ararat.

Human-Ape common ancestry from ERVs drives the nail in the coffin of monogenism

            I’ve often referred to the evidence for human-ape common ancestry (which alone falsifies monogenism), so I don’t want to go into detail. The argument here is simple and elegant: if we see exactly the same ‘genetic error’ in exactly the same place in the genomes of humans and apes, the odds of the same mistake occurring purely by chance are billions to one against. When we have many such errors, the odds are so small as to be effectively zero. The ‘genetic errors’ include broken genes (pseudogenes), mobile genetic parasitic DNA (retrotransposons) that simply copy and paste themselves randomly throughout the genome, and endogenous retroviral elements, which are evidence of past retroviral infection that has integrated into the host germline.
            One of my favourite papers is a seminal 1999 PNAS paper from two world-renown virologists, Welkin Johnson and John Coffin that employed endogenous retroviral data in humans and apes to construct a robust evolutionary family tree that was consistent with existing evolutionary trees. The existence of the same ERV elements at exactly the same place in human and ape DNA is best explained by a retroviral infection of a common ancestor of human and ape that became fixed in the common ancestor DNA, and passed down both lines like other genetic data. As the authors note:
Given the size of vertebrate genomes (>1 × 109 bp) and the random nature of retroviral integration, multiple integrations (and subsequent fixation) of ERV loci at precisely the same location are highly unlikely. Therefore, an ERV locus shared by two or more species is descended from a single integration event and is proof that the species share a common ancestor into whose germ line the original integration took place. Furthermore, integrated proviruses are extremely stable: there is no mechanism for removing proviruses precisely from the genome, without leaving behind a solo LTR or deleting chromosomal DNA. The distribution of an ERV among related species also reflects the age of the provirus: older loci are found among widely divergent species, whereas younger proviruses are limited to more closely related species.[12] (Emphasis mine)
The results, as I noted before, showed that the pattern of retroviral insertion gave evolutionary trees consistent with consensus trees:
The HERVs analyzed above include six unlinked loci, representing five unrelated HERV sequence families. Except where noted, these sequences gave trees that were consistent with the well established phylogeny of the old world primates, including OWMs, apes, and humans.[13] (Emphasis mine)
I’m not the only one to be impressed by this evidence (or this paper). New Zealand cell biologist and cancer researcher (and Christian) Graeme Finlay, in his book “Human Evolution: Genes, Genealogies and Phylogenies” (2013: Cambridge University Press) notes:
A 1982 study prepared the way (as far as I was concerned!) for the surprising answer [of when such retroviral DNA first entered the human genome]. A length of cloned human chromosomal DNA had been mapped on the basis of restriction enzyme-cutting sites (that provide sequence landmarks along the DNA). An equivalent piece of DNA cloned from the chimpanzee showed almost the same restriction enzyme-mapping sites, indicating that these lengths of cloned DNA were from the corresponding parts of the two genomes. But what is remarkable was that each of these segments of DNA overlapped the sequence of an ERV. This finding implied that the ERV in each of the two genomes was inserted at the same location. If indeed it was the same insert (same class of ERV, inserted in precisely the same site with the same target-site duplication, and lying in the same direction), then we would have to conclude that both species are descendants of the single progenitor in which this unique insert event occurred. This remarkable conclusion, reflecting the way in which shared proviruses establish the monoclonality of tumours, was forced on me by every instinct inculcated by cell biological experience.
But was the ERV indeed the same one in both species? The definitive answer could only come from DNA sequencing studies, and this pioneering work preceded the high-throughput sequencing revolution. DNA sequencing had not been performed on these cloned lengths of human and chimp genome. The answer was not available. However, this research held out the tantalising prospect that the sequencing of ERV integration sites in related species might provide the definitive answer to the question of whether humans and chimps are monoclonal (as a cell biologist might express it). The word monophyletic applies more appropriately to multiple species descended from one ancestor. The distribution of ERVs in the DNA of primate species could provide the ultimate statement on common descent.
Work published in 1999 [the Johnson and Coffin paper] settled the question of whether shared ERVs could demonstrate human and chimp descent from a common ancestor. This seminal study identified those primate species in which each of six ERVs was present – and defined insertion sites at single-base resolution. The data confirmed that each of these ERVs is shared by humans and chimps. Indeed, each ERV is shared not only by humans and chimps, but also by gorillas and more distantly related primate species. [14]

This is of necessity a short review of the evidence against monogenism, but is more than enough for the honest, disinterested reader to see that the human population has never been as small as two people. 


[1] Liu H et al “A Geographically Explicit Genetic Model of Worldwide Human-Settlement History” Am. J. Hum. Genet (2006) 79:230-237

[2] Li H., Durbin R “Inference of human population population history from individual whole-genome sequences.” Nature (2011) 475:493-496

[3] The 1000 Genomes Project Consortium “A global reference for human genetic variation” Nature (2015) 526:68-74

[4] Schiffels S., Durban R “Inferring human population size and separation history from multiple genome sequences” Nature Genetics (2014) 46:919-925

[5] Mallick S et al “The Simons Genome Diversity Project: 300 genomes from 142 diverse populations”  Nature (2016) 538:201-206

[6] Schiffels and Durban, p 206

[7] Geneticist Dennis Venema pre-empts this creationist attempt to evade the power of this evidence for minimum population sizes, “Ah, you might say, these studies require an estimate of mutation frequencies from the distant past. What if the mutation frequency once was much higher than it is now? Couldn’t that explain the data we see now and still preserve an original founding couple? Aside from the problems this sort of mutation rate would present to any species, we have other ways of measuring ancestral population sizes that do not depend on mutation frequency. These methods thus provide an independent way to check our results using allele diversity alone.” Venema, Dennis R.; McKnight, Scott. Adam and the Genome: Reading Scripture after Genetic Science. Brazos Press, 2017. p48-49.

[8] Temesa et al “Recent human effective population size estimated from linkage disequilibrium” Genome Res. (2007) 17: 520-526

[9] Venema D “Genesis and the Genome: Genomics Evidence for Human-Ape Common Ancestry and Ancestral Hominid Population Sizes” Perspectives on Science and Christian Faith (2010) 62:166-178

[10] Rosenberg A.A., and Feldman M.W. “The relationship between coalescence times and population divergence times” in Slatkin M, Veuille M (Eds) “Modern Developments in Theoretical Population Genetics” (2002: Oxford University Press)

[11] Excoffier L “Human demographic history: refining the recent African origin model” Curr Opin Genet Dev (2002) 12:675–682

[12] Johnson W.E., Coffin J.M. “Constructing primate phylogenies from ancient retrovirus sequences” Proc Natl Acad Sci USA. (1999) 96:10254-10260

[13] ibid, p 10259

[14] Finlay, p 34-35