Displacement Of The Neandertal Population Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Ever since the displacement of the Neandertal population by anatomically modern humans circa 30,000 years before present history has continued to mould and shape the populations that inhabit the British Isles, and by association their genetic identities. Figure 1 plots the historical events that have most contributed to the makeup of the modern day population. The earliest record of anatomically modern humans inhabiting the British Isles is the Red Lady of Paviland, the Last Glacial Maximum caused widespread depopulation of Northern Europe, the subsequent repopulation as the ice retreated marked the start of the Mesolithic era. The Neolithic revolution marks the spread of farming across the Isles and the significant increase in population that followed. More recently it was cultural transitions that were responsible for demographic change, The Celtic Revolution (500 B.C.E), Roman Invasion (43 C.E), Anglo-Saxon Invasion (500 C.E), Viking Invasion (800 C.E), and the Norman Conquest (1066 C.E.) have all had varying impacts across the different regions of Britain.

The extent to which these more recent cultural transitions were determined by migration remains unproved. Pre-1960 archaeological evidence representing cultural transition, such as changes in buildings, pottery, and burial rituals that are characteristic of a different population ,was accepted as prima facie evidence for a mass immigration of said population into the geographic location of the evidence. This view is rejected by the movement of New Archaeology in favour of the 'elite dominance model' (Renfrew, 1987) whereby the widespread adoption of a new culture could occur without the need for mass immigration but via trade or as a result of an influx of a ruling elite and their resulting influence over the native population. Both views have since varied in popularity as evidence emerges to support both, for example there is little evidence to suggest mass immigrations into Britain during the Norman Conquest or the Roman Invasion, thus appearing to support an elite dominance view, whereas by contrast there is evidence showing Viking settlement in Orkney Grampian and East Anglia.

Perhaps the most controversial cultural transition from British history is that from Romano-Britain to Anglo-Saxon Britain (circa 400 C.E.) and the extent to which migration was a factor. Unlike the Roman transition before it and the Norman transition after there is a profound lack of contemporary records regarding the Anglo-Saxon invasion and subsequent settlement. Much of the contemporary evidence of the period comes from just two writings, those of Gildas (540 C.E.) and Bede (731 C.E.), with both reporting a large scale immigration into Britain. Bede identified the Anglo-Saxon's as the descendants of three Germanic tribes (Figure 2):

The Angles, believed to have originated from Angeln

The Saxons, from Lower Saxony

The Jutes, from the Jutland Peninsula

Another potential source of origin for Anglo-Saxon migrants is Friesland as both it's geographical location and the similarities between the modern Frisian language and that of Old English (Nielsen 1985) make Friesland a suitable candidate for the source of invading Germanic tribes. Bede noted of the Angles that there entire nation came to Britain, leaving their former land empty. Archaeological evidence confirmed a marked increase in continental culture in Britain and the apparent desertion of Germanic settlements during this period (Esmonde-Cleary 1993) appearing to lay support for the claims made by Bede.

Figure 2 - Routes taken by the Angles, Saxon's and Jutes to Britain

However questions have been raised with regard to the historical accuracy of the contemporary data supporting the invasion hypothesis (Crawford 1997, Hamerow 1997), leading some authors to reject this model of large scale immigration/emmigration and propose a counter model for the transition whereby there is a continuity of the Romano-British population and that the sudden change to Anglo-Saxon culture is explained by rapid acculturation (Arnold 1984, Higham 1992) possibly due to the influence of small numbers of Germanic immigrants who, relative to the native population, were elite. Crucially, under this hypothesis one would not expect to see a significant Germanic contribution to the British gene pool.

In more recent years genetic studies have attempted to clarify such cultural transitions with the non-recombining portion of the Y chromosome and the control region of mitochondrial genome forming excellent markers with which to track larger scale past migrations and settlements. Previous studies of both Y chromosome and mitochondrial DNA variation have suggested that the population expansions of the Paleolithic and Neolithic eras have resulted in large scale clines (Torroni, Richards) Weale et al's 2001 Y-chromosome study on Anglo-Saxon migration revealed the distribution of Y haplotypes between Central England and Friesland was remarkably similar and that between Central England and Wales it was dissimilar suggesting a genetic barrier between the English and Welsh. Using population genetic models to explain this similarity they concluded the data provided evidence of a massive Anglo-Saxon male migration into Britain so as to contribute between 50-100% of the male population at the time. Capelli et al 2003 studied the distribution of Y haplotypes in the British Isles, Norway, Southern Denmark, Northern Germany and the Iberian peninsula. They observed a large degree of heterogeneity in the pattern of Continental input to the English gene pool across the Isles but estimated that Continental introgression into England ranged between 24.4 and 72.5% (mean 54.1%). In addition they found a marked similarity between the Y chromosomes of areas that did not not experience Continental input such as Wales and those of the Iberian peninsula. Even in areas of relatively high continental input this Iberian similarity was pronounced. Thomas et al 2006 equated that these high proportions of introgression to a migration in excess of 500,000 well above any estimated for population movements of the time (Heather 1991) and proposed this Germanic input could be explained by the existence of an Apartheid like social structure in England shortly after the Anglo-Saxon invasion whereby the higher economic and social status achieved by the continental immigrants results in an increased chance of reproductive success thus increasing the Continental contribution to the gene pool.

In the same way that the Y chromosome can trace the agnate lineage, the exclusively maternal inheritance of the mitochondrial genome combined with a rapid mutation rate and the genome not being highly conserved make it an ideal genealogical tool with which to trace the history of the maternal lineage. The purpose of this study is to repeat much of the methodology used by Weale et al 2002 but on mitochondrial DNA sequences so to evaluate the effect of the Romano to Anglo-Saxon cultural transition on matrilineal descent. Wilson et al 2001 analysed Y chromosome, X chromosome and mitochondrial data from the British Isles and several other populations and concluded that whilst the paternal lineage showed significant continental and Scandinavian input, the maternal lineage was largely influenced by a pre Anglo-Saxon cultural transition and has not had any significant input since. However the Wilson et al study did not compare English populations with the continent and did not address the impact of migration during cultural transitions. This study uses previously published mitochondrial DNA data from British and Continental populations in addition to novel simulation based computer modelling to discover the Anglo-Saxon influence on the maternal lineage of England, and whether this is consistent with the model of mass immigration proposed by Weale.

Materials and Methods

Data Collection

Mitochondrial DNA sequences were collected from published studies and databases (Mitomap, Mitochondrial Concordance, HmtDB, and EMPOP). The comparative dataset consisted of Z HVS-1 sequences from the following populations and sources: Great Britain (3687 from Sykes 2006 and 502 from Goodacre et al 2005), Germany (Upper Saxon Circle - 1200 from Pfeiffer et al 2001 and Lower Saxon Circle - 177 from Tetzlaff et al 2007), Friesland (77 from Wilson et al 2001), Denmark (201 from Mikkelsen et al 2009) Norway (323 from Helgason at al 2001) . Data was converted from the native format, expressed as differences from Anderson, into sequence (fasta) format using scripts written in the Python programming language (http://www.python.org/). All samples were aligned using ClustalX (www.clustal.org/), sequence lengths were modified using Se-Al (http://tree.bio.ed.ac.uk/software/seal/) so as to output sequences of uniform length 315bp, from positions 16050-16365 of Hyper Variable Segment I of the mitochondrial genome.

Regional Boundaries

The British data was further classified by regional area to allow for finer analysis, the regional boundaries used in this study are taken from those used in the Oxford Genetic Atlas Project (Sykes 2006) as this source was by far the largest contributor to the British dataset. The regional boundaries used by Sykes are shown in Figure X. In accordance with Weale et al 2001 to assess the extent of female Anglo-Saxon migration the suspected source and origins populations were amalgamated into Central England & East Anglia and Saxon Circle & Friesland respectively.

Statistical Analysis

The comparative dataset was collapsed into haplogroups and classified by population using FaBox (www.birc.au.dk/~biopv/php/fabox/). Analysis of Molecular Variance (AMOVA) calculations were used to quantify among population genetic variation (pairwise FST) and p-values for the exact test of sample differentiation (Raymond 1995) amongst the comparative dataset. In addition to those described more detailed statistical analysis was performed on the source and origin populations including Pairwise difference (both amongst and between the populations), number of Haplotypes, number of polymorphic sites, Haplotype Diversity, Nucleotide Diversity. All statistics were generated using Arlequin (Schneider 2000).

Population Modelling

Population models capable of accounting for the data patterns were explored using Bayesian inference of demographic and genetic parameters under population splitting, growth, and migration using the bayesian coalescence program BayeSSC (Excoffier 2000 ,Anderson 2005) to generate approximate posterior distributions of the parameters of interest given the observed data and prior distributions, unlike the Markov chian monte carlo approach used by Weale et al 2002 BayeSSC does not require explicit likelihood functons and requires far less computational time.

Three population models were explored, firstly that of a null hypothesis with regard to Anglo-Saxon migration en masse: that the genetic difference between modern populations in Central England/East Anglia and North Germany/Friesland could be explained solely under a model of population growth since these two populations split during the Neolithic and thus need not involve migration. The priors used in this model were as follows; 3 population demes were assumed, the two modern populations and their respective ancestral population, the modern female effective population sizes (N0) were conservatively estimated as 3.7*106 and 4.4*106 respectively. Two periods of exponential growth were assumed, the first occurring after the Upper Paleolithic colonisation of Europe (45000 YBP or 1800 generations before present assuming 25 years per generation), the second occurring in the Neolithic after the split between the ancestral population of Northern Europe (6100 YBP or 244 generations before present assuming 25 years per generation) resulting in the two descendant populations, A and B, that would eventually become the modern Central English and Saxon Circle populations.

The model explored various permutations of Upper Paleolithic and Neolithic population sizes (NUP and NN respectively). The respective exponential growth rates were calculated as follows:

r = ln[Nt/N0]/t

Where t is the time in generations, N0 is the modern effective population size, and Nt is the population size t generations ago. This null hypothesis model used two parameters, Upper Palaeolithic effective population size and Neolithic effective population size. A range of values for these parameters were obtained using the minimum and maximum values (Bouquet Appel) for each era, in total 16 values were used per parameter resulting in 256 variations of the null model that were each simulated to 5000 iterations, resulting in a total of 1.28*106 simulations. The results of each parameter variation were than compared to those observed in the statistical analysis of the dataset and the probability (P) of obtaining the observed dataset value calculated using the formula for a two tailed t-test as described by Voight et al 2005:

1 - 2 * | 0.5 * P |

Where P is the p-value from a one tailed t test. The two-tailed P values were then plotted according to their parameters. T-tests were performed on each of the observed statistics listed above in Statistical Analysis before performing the Voight method (Voight et al 2005) for combining non-independant variables in t-tests to calculate the probability (P-value) of obtaining all the observed statistics for the parameters in the null model.

Two further population models were then explored. The second model again assumed population growth since the Neolithic split only this time symmetrical constant background migration (from Central England to North Germany and from North Germany to Central England) since the Neolithic split was included as a parameter. Background migration rates per generation were explored using a range of values from impossibly low (0%) to impossibly high (5%) as a point of reference background migration rate in the European Economic Community over the past 30 years are estimated at 0.1% (http://www.homeoffice.gov.uk/rds/) an implausibly high value for pre air-travel eras. The third model added a migration event as a parameter. This migration even was an asymmetrical migration from Northern Germany (Saxon source) into Central England (Saxon sink). The scale of the migration event parameter ranged from 0% (no extra migration event occurred) to an implausibly high 100% (the entire population of North Germany migrated into Central England leaving North Germany empty). The first model was then slightly altered to correspond with these models, rather than using a range of 16 values for each of the NUP and NN parameters these parameters could take any value in range Min Ne to Max Ne (10 to 5000 and 1000 to 100000 respectively).

Each model was then simulated to 1*106 iterations and rejection model Approximate Bayesian Computation (ABC) (Beaumont et al 2002) performed on the resulting data. This algorithim compares the summary statistics of the observed data with each set of summary statistics for the simulated dataset, retaining the simulated data that closely matches the observed data and rejecting the rest. The retained parameter values are then adjusted using weighted local-linear multiple regression and a random sample generated to produce a marginal posterior distribution and a modal point estimate for each parameter. This data is then plotted, indicating which model best explains the observed summary statistics and estimates for parameter values that best fit the observed summary statistics within the model. These parameter values are then compared against known estimates for plausability.


As the focus of this study is Anglo-Saxon migration into England the Scottish populations are hereafter amalgamated under the population Scotland, the results of regional Scottish populations are included in the appendix under Supplementary Table X for reader interest.

Table 1 - Populations and Sample Sizes

Patterns of Genetic Differentiation

When analysing the genetic distance of the comparative dataset (Table 2) several patterns emerged, the most striking of which was the high degree of homogeneity amongst the populations of the British Isles, with the FST typically being between 2 to 3 orders of magnitude smaller than those observed by Weale. In stark contrast to Weale et al 2001 there was no marked difference in FST between adjacent English populations compared to those between British national borders; Central England/Wales (0.00064) and Northumbria/Scotland (-0.00106). It is immediately evident that in such a homogenous population the influences of any one cultural transition on the maternal lineage are likely going to be far more subtle than they were for the paternal lineage.

Table 2 - Genetic Distances and P-Values

Upper right triangle contains the P-Values of the Pairwise exact test for population as described by Raymond and Rousset (1995) based on haplotypic frequencies. An asterix denotes a significant value (P < 0.05). Lower left triangle contains AMOVA base FST ValuesAn initial principle co-ordinate (PCO) plot of the FST matrix (Figure 4) instantly revealed Norway to be a outlier, suggesting little Nordic input to the British gene pool, a further PCO plot with Norway omitted (Figure 5) allowed for a smaller scale with which to view the differentiation in the remaining populations. Here Friesland is an outlier suggesting only limited Frisian input, a curious anomaly given that in Weale et al 2002 Friesland was shown to be remarkably similar to Central England and a significant source of input to the gene pool of Central England, however it should be appreciated that whilst relatively speaking our results suggest a limited Frisian input, in absolute terms the difference in FST between Central England and Friesland is similar in both studies (0.004 in Weale et al 2002 compared to -0.00658 in this study) it is only the stark contrast in British population homogeneity that results in the differing implications with regard to Frisian input to the Central English gene pool.

Furthermore, of the continental populations in the comparative dataset, by far the most closely related to the English populations was the Saxon Circle population. The similarity with the Central English population being the most striking (FST = -0.00004). From Figure 5 it is evident that the Saxon Circle population has had a marked influence on the populations of both Central England and East Anglia to such an extent that the North Sea appears to have acted as less of a genetic barrier than the Anglo-Welsh border (FST = 0.00064). In the North of England and Scotland it is evident that there is a larger Danish influence to the gene pool (Figure 5) and also a slight Norwegian influence in Northumbria (Figure 4). It should be noted that the relatively high degree of differentiation between the Frisian and Irish populations when compared to the dataset may in part be due to the small sample sizes of these populations (n = 84 and 39 respectively).


The British data was further classified by regional area to allow for finer analysis, the regional boundaries used in this study are taken from those used in the Oxford Genetic Atlas Project (Sykes 2006) as this source was by far the largest contributor to the British dataset. The regional boundaries used by Sykes are shown in Figure X.

Literature Cited

Anderson CNK, Ramakrishnan U, Chan YL and Hadly EA (2005) Serial SimCoal: A population genetic model for data from multiple populations and points in time. Bioinformatics, 21, 1733-1734

Bramanti, B., Thomas, M. G., Haak, W., Unterlaender, M., Jores, P., Tambets, K. et al. (2009). Genetic discontinuity between local hunter-gatherers and central Europe's first farmers. Science, 326(5949), 137-140. doi: 10.1126/science.1176869

Capelli, C., Redhead, N., Abernethy, J. K., Gratrix, F., Wilson, J. F., Moen, T. et al. (2003). A Y chromosome census of the British Isles. Curr Biol, 13(11), 979-984.

Excoffier L, Novembre J and Schneider S (2000) SIMCOAL: a general coalescent program for simulation of molecular data in interconnected populations with arbitrary demography. J. Hered., 91, 506-509.

Goodacre, S., Helgason, A., Nicholson, J., Southam, L., Ferguson, L., Hickey, E. et al. (2005). Genetic evidence for a family-based Scandinavian settlement of Shetland and Orkney during the Viking periods. Heredity, 95(2), 129-135. doi: 10.1038/sj.hdy.6800661

Helgason, A., Hickey, E., Goodacre, S., Bosnes, V., Stefansson, K., Ward, R. et al. (2001). mtDna and the islands of the North Atlantic: estimating the proportions of Norse and Gaelic ancestry. Am J Hum Genet, 68(3), 723-737. doi: 10.1086/318785

Mikkelsen, M., Sørensen, E., Rasmussen, E. M., & Morling, N. Mitochondrial DNA HV1 and HV2 variation in Danes. Forensic Science International: Genetics, In Press, Corrected Proof. doi: DOI: 10.1016/j.fsigen.2009.07.007

Miller, K. W., & Budowle, B. (2001). A compendium of human mitochondrial DNA control region: development of an international standard forensic database. Croat Med J, 42(3), 315-327.

Pfeiffer, H., Forster, P., Ortmann, C., & Brinkmann, B. (2001). The results of an mtDNA study of 1,200 inhabitants of a German village in comparison to other Caucasian databases and its relevance for forensic casework. Int J Legal Med, 114(3), 169-172.

Sykes B (2006) Blood of the Isles: Exploring The Genetic Roots of Our Tribal History. Bantam Books

Tetzlaff, S., Brandstatter, A., Wegener, R., Parson, W., & Weirich, V. (2007). Mitochondrial DNA population data of HVS-I and HVS-II sequences from a northeast German sample. Forensic Sci Int, 172(2-3), 218-224. doi: 10.1016/j.forsciint.2006.12.016

Thomas, M. G., Stumpf, M. P., & Harke, H. (2006). Evidence for an apartheid-like social structure in early Anglo-Saxon England. Proc Biol Sci, 273(1601), 2651-2657. doi: 10.1098/rspb.2006.3627

Weale, M. E., Weiss, D. A., Jager, R. F., Bradman, N., & Thomas, M. G. (2002). Y chromosome evidence for Anglo-Saxon mass migration. Mol Biol Evol, 19(7), 1008-1021.

Wilson, J. F., Weiss, D. A., Richards, M., Thomas, M. G., Bradman, N., & Goldstein, D. B. (2001). Genetic evidence for different male and female roles during cultural transitions in the British Isles. Proc Natl Acad Sci U S A, 98(9), 5078-5083. doi: 10.1073/pnas.071036898

Raymond, M., and F. Rousset (1995). An exact test for population differentiation. Evolution 49:1280-1283.

Esmonde-Cleary, A. S. (1993). Approaches to the differences between Late Romano-British and Early Anglo-Saxon archaeology. Anglo-Saxon Stud. Archaeol. Hist. 6:57-63.

Crawford, S. 1997. Britons, Anglo-Saxons and the Germanic burial ritual. Pp. 45-72 in J. Chapman and H. Hamerow, eds. Migrations and invasions in archaeological explanation. BAR International Series 664, Oxford, UK

Hamerow, H. 1997. Migration theory and the Anglo-Saxon ''identity crisis.'' Pp. 33-44 in J. Chapman and H. Hamerow, eds. Migrations and invasions in archaeological explanation. BAR International Series 664, Oxford, UK

Arnold, C. J. 1984. From Roman Britain to Saxon England. Croom Helm, London.

Higham, N. 1992. Rome, Britain and the Anglo-Saxons. Seaby, London.

Torroni, A., H. Bandelt, H. D'Urbano et al. (11 co-authors).1998. mtDNA analysis reveals a major Late Paleolithicpopulation expansion from southwestern to northeasternEurope. Am. J. Hum. Genet. 62:1137-1152.

Richards, M., V. Macaulay, E. Hickey et al. (37 co-authors).2000. Tracing European founder lineages in the NearEastern mtDNA pool. Am. J. Hum. Genet. 67:1251-1276

Beaumont, M. A., Zhang, W., & Balding, D. J. (2002). Approximate Bayesian computation in population genetics. Genetics, 162(4), 2025-2035.

Voight, B. F., Adams, A. M., Frisse, L. A., Qian, Y., Hudson, R. R., & Di Rienzo, A. (2005). Interrogating multiple aspects of variation in a full resequencing data set to infer human population size changes. Proc Natl Acad Sci U S A, 102(51), 18508-18513. doi: 10.1073/pnas.0507325102