Deep maternal ancestry and mtDNA

In February 2014, I did a series of posts on my deep maternal ancestors, identified through a test of mutations on my mitochondrial DNA (mtDNA) which is inherited only from the mother. This test was carried out by, who have since discontinued tests of mtDNA and Y-chromosome DNA. Costs of DNA tests have dropped dramatically since then, and late last year I ordered an mtDNA test from FamilyTreeDNA ( which carried out a full sequencing of the mitochondrial DNA.

As well as the DNA that makes up the chromosomes in the nuclei of our cells, we also have another type of DNA called mitochondrial DNA (mtDNA). The mitochondria are organs located outside the cell nucleus which convert sugars into energy.  Mitochondria have a small circular loop of DNA, containing only approximately 16,569 base pairs in humans. The circular mtDNA is similar to the DNA of bacteria, and it is thought that mitochondia evolved from symbiotic bacteria that were once free living.

In humans, as in other higher organisms, a DNA molecule consists of two strands that wrap around each other in the well known double helix structure. Each strand contains a linear arrangement of four different bases: adenine (A), thymine (T), guanine (G), and cytosine (C).  Specific sequences of three bases make a DNA “word” which codes for an amino acid. A string of DNA “words” strung together in a sequence provides the instructions to make a particular protein. The base-pairs on the circular mtDNA loop are numbered from 1 to 16,569 and different portions of the loop have been given different names as shown in the following diagram. The first and second portions, called hypervariable control region 1 (HRV1) and hypervariable control region 2 (HRV2) are regions of the mtDNA that accumulate mutations (changes in base-pairs) relatively quickly and thus tend to be hyper-variable between people who are not closely related. The third portion, the coding region, accumulates far fewer changes and contains the base-pair sequences for mitochondrial genes [1].

mtDNA molecule (Credit: Debbie Parker Wayne)

One type of mutation, called a single nucleotide polymorphism (SNP), is a single base pair in a DNA sequence that has been replaced by a different base pair. My mtDNA test results identified 68 SNPs, 13 in HRV1, 17 in HRV2, and 38 in the coding region through comparison of my base sequences with those of the Reconstructed Sapiens Reference Sequence (RSRS). The RSRS is a recent effort to reconstruct a single ancestral genome for all living humans, using both a global sampling of modern human samples and samples from ancient hominids. It was introduced in early 2012 as a replacement for the rCRS (revised Cambridge Reference Sequence).

Mitochondrial DNA is inherited from the mother alone, rather than being inherited from the father and the mother. Additionally, recombination (or crossing over) does not occur in mitochondrial DNA (mtDNA). For both of these reasons, the sequence of mitochondrial DNA stays largely the same over generations, and thus is a useful tool for looking at maternal ancestry.

Over a period of nearly 180,000 years, SNPs have steadily accumulated on different human mtDNA molecules being passed down from mothers to their children. SNPs thus provide a cumulative dossier of our own maternal prehistory. We can use these mutations to reconstruct a genetic tree of mtDNA, because each new SNP in a prospective mother’s ovum will be transferred in perpetuity to all her descendants down the female line.

When a group of people share similar SNPs, they are part of the same “haplogroup”. For example, over 95% of native-born Europeans fit into seven haplogroups, which in turn derive from an older haplogroup that arose in the Middle East. Other regions of the world are associated with different haplogroups. Each of these groups trace their maternal ancestry back to just one woman, the common maternal ancestor of everyone in her haplogroup, or clan. Not everyone in the same clan has exactly the same mtDNA, because DNA accumulates additional mutations gradually over the generations. However, everyone in the clan shares a set of common mutations, which are the signature of the mtDNA of their founding maternal ancestor.

By averaging the numbers of mutations found in the mtDNA of modern members of a haplogroup, and knowing the average mutation rate for mtDNA, as well as the dating of ancient human remains whose mtDNA has been sequenced,  it is possible to estimate how old each clan is, or in other words, when the common maternal ancestor of the clan lived (see below). By studying features of the geographical distribution of their present-day descendants, as well as the locations of ancient human remains whose mtDNA has been sequenced, it is possible to work out where they most likely lived as well. Generally speaking, the likely geographic origin for a clan is not necessarily the place where it is most common today, but the place where it is the most varied.

It is thus possible to trace migration routes by observing the branching points in an ancestral map containing all known haplogroups (see map below). Mitochondrial DNA in humans of African origin show the most diversity in the world. This supports the concept that ancient humans first existed in Africa and stayed in Africa for thousands of years.  The humans who left Africa around 70,000 years age took two major routes, to Asia (haplogroup M) and to Europe (haplogroup N).

Migration routes of human beings dating back to 170,000 years ago. All humans originated in Africa and migrated out, branching into the two main out of Africa haplogroups, M and N. Individuals in haplogroup M headed west to Asia and later to the Americas, while haplogroup N moved into Europe.

The clan mothers were not the only people alive at the time, of course, but they were the only ones to have direct maternal descendants living right through to the present day. The other women around, or their descendants, either had no children at all or had only sons, who could not pass on their mtDNA. And, of course, the clan mothers had ancestors themselves. Everyone alive on the planet today can trace their maternal ancestry back to just one woman, the founder of haplogroup L. According to recent studies, she lived in Africa nearly 180,000 years ago and is known as “Mitochondrial Eve” (see my previous post  mitochondrial eve )

Age Estimates

The age estimates for haplogroup founders in my 2014 posts were based on estimating average mutation rates for mtDNA SNPs, assuming an average constant mutation rate [2, 3]. There have been several more recent studies that have updated these estimates using not only samples from living humans but also analyses of mtDNA from ancient humans.

My previous post on mitochondrial Eve used dates from Behar et al. 2012 [4]. Comparing these with the dates given on the FamilyTreeDNA site and on the site  snpTracker, as well as as two other more recent papers [5, 6], there is considerable variation in some of these. I revised the date for mitochondrial Eve based on the average of the dates from Behar et al [4] and Fu et al [5], and the dates for L3 and U based on the average of the dates from Behar et al [4], Fu et al [5] and Soares et al [6]. Dates from these three papers were reasonably consistent. Dates for other haplogroups from Behar et al were adjusted accordingly. 90% uncertainty ranges were estimated using the relative uncertainty ranges Behar et al. [4]. Because of the random nature of individual mutations (they may not occur at exactly the average time to next mutation), there is an uncertainty range around dates (which is also statistically estimated).

The following map summarizes the revised timeline for the migration of the ancestors of maternal haplogroup U5 out of Africa.

Ancestral migration path of maternal ancestors for haplogroup U5

In following posts, I will update the previous posts summarizing my maternal haplogroup ancestors and placing them in the context of the human expansion out of Africa and across Europe, as well as the context of the ice ages and the evolution of human cultures.


[1] Blaine T. Bettinger (2016). The Family Tree Guide to DNA Testing and Genetic Genealogy. Family Tree Books: Cincinnati.

[2] Behar D, Villems R, Soodyall H, Blue-Smith J, Luisa Pereira L, Ene Metspalu E, Rosaria Scozzari R, Heeran Makkan H, Shay Tzur, David Comas, Jaume Bertranpetit, Lluis Quintana-Murci, Chris Tyler-Smith, R. Spencer Wells, Saharon Rosset (2008). The Dawn of Human Matrilineal Diversity. American Journal of Human Genetics, 82(5):1130-1140.

[3] Soares P, Ermini L, Thomson N, Mormina M, Rito T, Röhl A, Salas A, Oppenheimer S, Macaulay V, Richards MB (2009). Correcting for purifying selection: an improved human mitochondrial molecular clock American Journal of Human Genetics, 84(6):740-59.

[4] Behar D, van Oven M, Rosset S, et al. A “Copernican” Reassessment of the Human Mitochondrial DNA Tree from Its Root. Am J Hum Genet. 2012;90(5):936. doi:10.1016/j.ajhg.2012.04.007 Open ArchiveDOI:

[5] Fu Q, Mittnik A, Johnson PLF, et al. A revised timescale for human evolution based on ancient mitochondrial genomes. Curr Biol. 2013;23(7):553–559. doi:10.1016/j.cub.2013.02.044

[6] Soares S, Alshamali F, Pereira JB, Fernandes V, Silva NM, Afonso C, Costa MD, Musilová E, Macaulay V, Richards MB, Černý V, Pereira L, The Expansion of mtDNA Haplogroup L3 within and out of Africa, Molecular Biology and Evolution, Volume 29, Issue 3, March 2012, Pages 915–927,

Mitochondrial Eve – an update

A recently published paper in Nature (Oct 18) has analysed the mitochondrial DNA of 1,200 indigenous Africans living in the southern part of Africa and identified the ancestral homeland of all humans alive today, the place where mitochondrial Eve lived nearly 200,000 years ago. More on that below, but first some background.

In February 2014, I did a series of posts on my deep maternal ancestors, identified through a test of mutations on my mitochondrial DNA (mtDNA) which is inherited only from the mother. These mutations allowed me to track back through time to mitochondrial Eve, the single woman from whom all humans alive today descended through their female line (mother to mother to mother….).  Specific mutations on the mtDNA define maternal haplogroups, and the founder of a given haplogroups is the specific individual woman in which the defining mutation occurred. All members of a given haplogroup trace their maternal ancestry back to this founder.

DNA tests have become much less expensive, and can include much more detailed testing. In the last three months, I’ve redone a test on my mtDNA and also on my Y DNA, which is inherited only down the male line (father to father to father….). I am still digesting the results of these tests, and will post on them in the near future.  One of the first things I discovered was that the dates associated with haplogroup founders have been revised over time, and as more and more test results are available, and that the terminology used for identifying haplogroups has also evolved.  I also came across very recent research which has pinned down the location where mitochondrial Eve lived, as well as revised estimates of the time period in which she lived.

Haplogroup U5 – the oldest of seven native European haplogroups

My mtDNA haplogroup is U5, the oldest of the seven native European haplogroups. Haplogroup U most likely arose in the Near East, and spread into Europe in a very early expansion, giving rise to seven native European haplogroups, including U5. The presence of haplogroup U5 in Europe pre-dates the last ice age and the expansion of agriculture in Europe. Today, about 10% of modern Europeans are the direct maternal descendants of the founder U5 woman, who has been given the nickname Ursula*. They are particularly well represented in western Britain and Scandinavia.

Ancestral migration path of maternal ancestors for haplogroup U5

Haplogroup U in turn is descended via haplogroups R and N from haplogroup L3, which is associated with a migration of humans out of Africa around 70,000 to 50,000 years ago. The dominant theory of human origins, the “recent African origin” theory, proposes that all modern non-African populations are substantially descended from populations of H. sapiens that left Africa after during that time period. H. sapiens most likely developed in Africa between 300,000 and 200,000 years ago, and there were at least several “out-of-Africa” migrations of modern humans, possibly beginning as early as 270,000 years ago. These early dispersals may have died out or retreated, although some paleoanthropologists argue that they possibly interbred with various other local hominid species and with later humans from “recent-out-Africa” and it just so happens that all the maternal lineages trace back to “recent-out-Africa”. Of all the lineages present in Africa, only the female descendants of Lara*, founder of the L3 haplogroup, are found outside Africa. If there had been several migrations, one would expect descendants of more than one lineage to be found.  Of course, all this could be upturned if descendants of other African lineages are found outside Africa, and can be traced back to earlier migrations.

Mitochondrial Eve (haplogroup L)

Mitochondrial Eve (mt-Eve) is a member of Haplogroup L and lived just before the divergence of macro-haplogroup L into L0 and L1–6 (see diagram below). Today the haplogroup L0 and its offshoots are found mainly in southern and eastern Africa, with particularly high frequencies among the San people (bushmen) of Botswana, Namibia and other countries of southern Africa.

Haplogroup L1 is found in West and Central sub-Saharan Africa. The descendants of haplogroup L1 are also African haplogroups L2 and L3, the latter of which gave rise to all non-African haplogroups.

Phylogenetic tree for mtDNA Haplogroup L, commencing with mitochondrial Eve, the most recent common maternal ancestor (MRCA) of all humans.

A recent paper by Chan et al. in Nature (October 2019) [1] analysed the genomes of more than 1,200 indigenous Africans living in southern Africa and claim to have identified precisely where and when the L haplogroup split into L0 and L1 and when these groups migrated from their homeland.

Chan et al. identified this homeland as Makgadikgadi, a vast wetland some 120,000 square kilometers in area, or roughly twice the area of Lake Victoria, Africa’s largest lake today. Mitochondrial Eve and her descendants lived in this region for about 30,000 years (from 200,000 to 170,000 years ago) before the L0 lineage split into its first subgroup. Today, Makgadikgadi is one of the largest salt flats in the world. Climate models suggest that, 200,000 years ago, it was a fertile oasis.  The map  shows the overall location of Makgadikgadi in southern Africa, and the following map shows  a more detailed view.

Satellite view of the Makgadikgadi salt pans. This area is located about 250 km south of Victoria Falls close to the borders of Zambia, Zimbabwe and Botswana.

Chan et al [1] date the deepest rooting L0 branch to 200,000 years ago (with 95% confidence interval 165,000 – 240,000 years ago).  I have reviewed the most recent comprehensive dating of maternal haplogroups and found that the dates in Fu et al (2012)  [2] and Behar et al [2013] were in reasonably good agreement.  I have used dates from Behar et al, which give a date of  176,700 years ago (confidence interval ± 11,300 years) for mitochondrial Eve, and 136,300 (± 11,700) years ago for L1. This is substantially earlier than the date of the recent out-of-Africa dispersal of L3 around 65,000 years ago.

The Okavango delta, in north-west Botswana, looks very similar to how Makgadikgadi would have looked 170,000-200,000 years ago. Credit: Diego Delso, CC BY-SA 4.0,

Migrations from the Makgadikgadi homeland

The Makgadikgadi wetlands were large, wet, and lush with vegetation. They would have provided an ideal home for wildlife and for early humans, including mt-Eve. So why did some migrate?  Around 130,000 years ago, there was a major climatic shift associated with the end of the penultimate glacial period. This led to higher rainfall and created “green corridors” leading to the northeast and to the southwest.  In particular, it appears that the ancestral founder of the L1 haplogroup lived around 136,000 years ago among a group that had migrated north into Zambia, and by around 70,000 years ago her descendents had made their way north to the horn of Africa, where Lara (L3 haplogroup founder live).

The “green corridors” proposed by Chan et al [1] helped lead humans out of the ancestral homeland

Chan and his group have extrapolated the likely location of mt-Eve’s homeland from the present-day distribution of the L haplgroup in Southern Africa, and it is always possible that future data may lead to revisions of this conclusion. However, multiple sets of evidence lead to the conclusion that mt-Eve was among the ancestors of the San people of southern Africa, although of course we likely will never know for sure exactly where she lived. And this was not the only ancestral human homeland. Y-DNA evidence suggests that Y-Adam lived in West Africa in a time period even further in the past (this will be subject of a future post) and of course, there may be other ancestral homelands associated with the many other ancestral lines than the purely maternal and paternal.

The San people of southern Africa have one of the most oldest maternal DNA lineages on Earth. They share the Haplogroup L with mitochondrial Eve who lived in northern Botswana nearly 200,000 years ago.

* Bryan Sykes in his 2001 book The seven daughters of Eve gave names to each of the women who founded the seven native European haplogroups, and also names to some of their ancestral haplogroups. He chose names that began with the letter by which the haplogroup was identified. Oppenheimer (The Origins of the British: A Genetic Detective Story, 2006) followed this example and also gave names to both mtDNA and Y-DNA haplogroups. To emphasise that the maternal clan founders were real individuals, who were my ancesters, I have used these names and given my own names to the more recent subgroup founders.


[1] Chan EKF, Hardie RA, Petersen DC, Beeson K, Bornman RMS, et al. (2015) Revised Timeline and Distribution of the Earliest Diverged Human Maternal Lineages in Southern Africa. PLOS ONE 10(3): e0121223.

[2] Fu Q, Mittnik A, Johnson PLF, et al. A revised timescale for human evolution based on ancient mitochondrial genomes. Curr Biol. 2013;23(7):553–559. doi:10.1016/j.cub.2013.02.044

[3] Behar D, van Oven M, Rosset S, et al. A “Copernican” Reassessment of the Human Mitochondrial DNA Tree from Its Root. Am J Hum Genet. 2012;90(5):936. doi:10.1016/j.ajhg.2012.04.007
Open ArchiveDOI: