|Chapter 19 - MtDNA|
|“… it is also a good rule not to put overmuch confidence in the observational results that are put forward until they have been confirmed by theory."|
|Sir Arthur Stanley Eddington, British astronomer and physicist|
“Eve” is a metaphor that the afrocentrists have given to our ancestral mother, who they believe lived in Africa about 150,000 ya (Shreeve, 2006), and from whom all living humans derived their mtDNA. “Eve” was not a single woman, however, since at least a thousand breeding pairs would be needed for a viable population. 1 According to the afrocentrists, all the women in that founding population either had the same mtDNA or, if they had different mtDNA, did not have daughters.
As support for Eve living in Africa, and for her date of 150,000 ya, afrocentrists point to studies of mtDNA in living people. Cells were collected from people all over the planet and were analyzed to determine the A-C-G-T base sequences (see Appendix) in their mtDNA; people within each geographically separated population tend to have many of the same A-C-G-T sequences, but those sequences are different in other populations. 2 For example, at a particular location (“locus”) on an mtDNA string, Europeans may have an A while Asians have a T. Differences in the A-C-G-T bases at a locus are called “SNPs” (single nucleotide polymorphisms). 3 Now the scientists had to decide what was the original base (A, T, C, or G) at each locus and which base (A, T, C, or G) is the mutation; the population with the original base was then be presumed to be the older, ancestral population from which all other populations descended. However, as we shall see, that reasoning may not be valid.
So the scientists programmed a computer that created millions of “trees,” with different populations at the bottom and on the various branches, based on “A’s” changing to “T’s” and “T’s” changing to “A’s” and so on. The assumption was made that the “correct” tree, that showed the actual changes that occurred in the bases over at least tens of thousands of years, would be the simplest tree, the most “parsimonious” tree. 4 The computer compared all these different trees and picked out the simplest tree and, low and behold, it was the tree with the Africans at the bottom, just as OoA had predicted.
The tree in Figure 19-1 (from Wikipedia) is imposed on a map of the world to show human migrations (black arrows) under the OoA theory. The blue lines represent the boundaries of areas covered in ice or tundra during the last ice age. The colors in the list on the right of the map are for the circles and the numbers for the colors in that list give the number of thousand years BP, i.e., before 1950. The letters and numbers inside the small white circles are for the groups of mtDNA alleles (“haplogroups,” see next chapter) that people living in those areas have. 5
The tree begins with an upside-down Africa in the tail of this weird-looking bird, then spreads to Asia (body) and Europe (left wing – orange circle), down to Australia (foot – red circle) and across the Bering Strait to North America (neck- blue circle) and South America (head – green circle). Unfortunately, the bird didn’t fly because the biologists who did the calculations were not mathematicians and, when a mathematician checked their work, he flunked them. The OoA tree was not the simplest tree. In fact, there were over a billion parsimonious trees. 6
So the mtDNA analysis does not show that Eve was an African. Can it at least tell us how long ago Eve lived? Since scientists now have all these mtDNA sequences and know how many SNPs there are, if they can assume that (1) every mutation that has occurred in Eve’s mtDNA is represented by a SNP in the data they have and (2) the mutation rate is constant (i.e., one mutation every X years), then they can easily calculate how long ago Eve lived, the “coalescence” date. 7 But are those two assumptions reasonable?
As to the first assumption, there are several reasons why the number of SNPs observed may be greater than the number of mutations that have actually occurred. Occasionally, during fertilization, the tail of a sperm will enter the egg along with its head, thereby adding the father’s mtDNA to the mother’s and possibly ending up in her daughter. (Hagelberg, 2003). If the father’s mtDNA is different from the mother’s, the scientists may count those differences as additional mutations, making the coalescence date seem farther in the past than it really was. Also, some of our male ancestors may have interbred with a female of another subspecies of Hs. If the daughters were accepted into our lineage, the scientists would count these additional SNPs as mutations and conclude that the coalescence date occurred much farther in the past than it did.
The number of SNPs may also be less than the number of mutations that have actually occurred. A mutation may occur, then later a second mutation may occur at the same location that reverses the first mutation, for example, A→T, then later T→A. The scientists don’t see any SNP at that location and they count no mutation, when really two mutations occurred, and therefore the coalescence date is older than they think it is. Also, two or more mutations may have occurred at the same site. Suppose A→T→G. All the scientists see is an A→G, so they count only a single mutation when there were really two mutations, and they again think that the coalescence date is more recent than it actually was.
Scientists have obtained ancient animal mtDNA 8 from fossil bones and teeth and date those bones by chemical and physical means. They can compare that mtDNA to mtDNA obtained from living descendants of those animals and count the number of SNPs. After adjusting as best they can for all the possible sources of error mentioned above, they divide the number of mutations by the number of years, which gives them the mutation rate, the number of mutations per year. They can then take the number of mutations in all living humans (as estimated from the number of SNPs), divide by the animal mutation rate and determine when all those humans started out with the same mtDNA (i.e., the coalescence date, the date that Eve lived).
But even if the number of SNPs is correctly adjusted for all the possible sources of error described above, the second assumption, that mtDNA mutates at a constant rate, must still be made. If, for example, hundreds of thousands of years go by and the mtDNA does not mutate at all and then there is a shower of cosmic rays or a volcano spews mutagens into the atmosphere, causing a large number of mutations, the mtDNA clock is not going to be accurate because it will be slowing down and speeding up. 9 And, when fossil bones are used to determine the mutation rate, additional assumptions must be made. The humans who lived at the time of the fossils and those who lived today were not genetically the same and may not have had the same resistance to mutations. After the Industrial Revolution, thousands of additional mutagens that never before existed were spewed into the atmosphere and the drinking water, so since about 1750 there may have been a higher number of mutations, making the date for Eve appear older than it was.
For these reasons, until technical problems are overcome, the mtDNA data cannot be relied upon for either the location of Eve or her date. 10 If the computer-generated tree made by afrocentrists does not prove that Eve lived in Africa, or even reliably when she lived, is it nevertheless possible to use the mtDNA data in another way to find out where she lived?
|Figure 19-3||Figure 19-4|
If both of the assumptions made by the afrocentrists (i.e., Eurasians and Africans are in the same lineage and more variation = older) are wrong and Eurasians did not evolve from Africans, but from a common ancestor with Africans (LCA), and the greater variation in African alleles is not due to their greater age, but to the infusion of DNA into Africa from multiple non-African hominoids who migrated there, 14 then the tree would look like Figure 19-4 (omitting intermediates). In Figure 19-4, the Eurasians did not descend from Africans and are not younger than Africans; Eurasians just gave Africans some alleles from time to time, adding to the variation in Africa.
As we shall see in Section IV, the tree of Figure 19-4 is more complicated, but it explains much more than does the OoA tree.
Table of Contents
1. (Harpending, 1998a). There is also a “50/500” rule of thumb, that at least 50 individuals are needed to begin a new population and at least 500 to keep it going for a long time. Back
2. OoA postulates that racial differences began only 65,000 ya, but some mtDNA differences between today’s Africans and Eurasians are older than that, as we shall see, which is not consistent with OoA. Back
3. SNPs that occur in less than 1% of a population are ignored for the purpose of establishing descent as they are considered random. About 90% of human genetic variations are SNPs; they occur every 100 to 300 nucleotides. (Human Genome Project Information, “SNP Fact Sheet”). Back
4. That assumption has been questioned because evolution does not always proceed straightforwardly. Also, there are many problems defining “parsimony” because the time between A-C-G-T changes is not known or considered and changes that took a very long time or a very short time may be incompatible with some trees. Also, the geographical locations where the A-C-G-T changes occurred is not known or considered and some of those locations may be too far away from the next step in the tree; the parsimony techniques gives equal weight to all changes, but some changes were no doubt much more important than others and critical changes must be in the right place on the tree, even if the tree is not parsimonious. (Schwartz, 2005, pp. 179-181). Back
5. Individuals who have the same haplogroup have interbred and are related. Thus, people in the M haplogroup in Australia are genetically close to the people in the M haplogroup in India. The Asians in the A, B, C, and D haplogroups are related to the Amerindians; the Ainu in Japan are also in haplogroup B. The X haplogroup in both Europe and in some Amerindians shows a relationship between them. Back
6. Henry Gee, a member of the editorial staff for the journal, Nature, described the studies as "garbage." Gee calculated that the total number of potentially correct parsimonious trees is somewhere in excess of one billion. (Gee, 1992). In a letter to Science, Mark Stoneking (one of the original researchers) acknowledged that the theory of an "African Eve" had been invalidated. ("African Eve theory takes a step back," New Scientist, Feb. 15, 1992). Back
7. The coalescence date is not necessarily the date that our species, Hss, began, however, though it may be. Back
8. There are thousands of copies of mtDNA in a cell and only one copy of nuclear DNA, so the chances of finding mtDNA preserved in old bones is much greater. Back
9. For example, a mutation may occur that is almost neutral and survives for thousands of years with few variations occurring in it. If the environment changes, e.g., a new disease, a different climate, that mutation may become vital and spread rapidly throughout the population, greatly increasing the number of people who have it and the number of variations in it. Back
10. “We call into question the use of mtDNA for studies of human evolution.” (Curnoe, 2003). Back
11. “... the gene pool in Africa contains more variation than elsewhere, and the genetic variation found outside of Africa represents only a subset of that found within the African continent. From a genetic perspective, all humans are therefore Africans, either residing in Africa or in recent exile.” (Pääbo, 2001). Back
12. “A surprising prediction of introgression [introducing new alleles by interbreeding] is that many genes may have a higher allelic diversity attributable to archaic introgression in Africa, not Eurasia.” (Hawkes, J., 2006). Also, “mtDNA diversity is essentially unpredictable and will, in many instances, reflect the time since the last event of selective sweep, rather than population history and demography.” (Bazin, 2006). I.e., when a mutation is positively selected, nearby alleles “hitchhike” along with it, so that as the mutation spreads, so do the hitchhiking alleles, thereby reducing variation in the genes of those alleles.
There is evidence besides greater diversity that afrocentrists could use to support their conclusion that Africans are older than Eurasians, but the afrocentrists do not rely on it, probably because it is a great embarrassment to them. Living Africans have alleles that chimpanzees and gorillas have, but Eurasians do not have. (Deka, 1995). This fact may, however, show not that Africans are older than Eurasians, but that they did not evolve as much as Eurasians – a population that begins in the tropics and stays there will not evolve as much as a population that begins in the tropics and slowly moves north into a temperate zone. Back
13. But if Eurasians did not come from a sub-population of Africans and are older, why do Eurasians have less variation? The answer is given in the next chapter. Back
14. “In Africa three races have intermingled to a certain extent with the negro; the Libyans (Berbers: q.v.) in the Western Sudan; and the Hamitic races (q.v.) and Arabs (q.v.) in the east.” ("Negro," 1911 Encyclopedia Britannica.) There has been so much infusion of non-Africans into Africa that non-African traits can be found even in fossils in southern Africa. (Chap. 26). Back