Where did I come from?

Testing the Hypotheses

The two hypotheses of human origins make different predictions about the evolutionary relationships among geographically separated groups of humans. Wilson and his colleagues tested the hypotheses by comparing their observations with the predictions. Support for a hypothesis is found when the observations match the predictions of that hypothesis.

Multi-regional Hypothesis
The Multi-regional Hypothesis has two important aspects: the idea that archaic humans (Homo erectus) left Africa and dispersed into the different geographic regions a very long time ago (about 1 million years ago), and that there was sufficient genetic mixing (gene flow) between populations to maintain a single species of humans. This means that we should observe deep genetic differences between the different geographic regions. The common genetic ancestor of modern humans should be very old. Futher, the African populations are not unique and should have patterns of genetic variation similar to other regions.

When the genetic relationships are estimated, we might observe a phylogenetic tree similar to the following, where each colour represents a different region.

Recent African Origin Hypothesis
The Recent African Origin Hypothesis implies that genetic variation among African samples will be different from that observed within other geographic regions. There should be a deep divergence between two groups of Africans, those related to the ancient Africans who left the region and gave rise to modern humans elsewhere and those whose ancient relatives did not leave Africa. Consequently the common ancestor of modern humans must be among the ancestors of modern Africans. The recentness of the origin of modern humans means that the common ancestor will have occurred more recently than under the Multi-regional hypothesis. If ancient Africans dispersed beyond that region, then all of the non-African regions will show relatively recent relationships with modern Africans.

Performing the Test by Inferring Genealogies from Mitochondrial DNA
Two pieces of evidence are needed to decide between the hypotheses: an estimate of the evolutionary relationships of modern humans from different geographical regions, and an estimate of the time of the most recent common ancestor of modern humans.

To do this, Wilson and his colleagues needed to use methods with which they could get relatively unambiguous estimates of human evolutionary relationships. The results needed to be reliable, that is they needed to be unaffected by other factors. They also needed to be precise, that is they needed to enable Wilson and colleagues to be able to distinguish between different proposed relationships.

The approach chosen was to estimate the relationships of people from different geographic regions by studying the variation in their mitochondrial genome. Mitochondria have several features that make them valuable for studying evolutionary relationships:

• There is a lot of material to work with - Most cells in your body contain mitochondria, structures where nutrients are turned into chemical energy. Each mitochondrion contains a small DNA genome containing about 16,000 base pairs. Each cell contains thousands of mitochondria, but only one nucleus.
• Mitochondria have high rates of evolution - The rate of evolution is like a microscope lens. A faster rate enables you to distinguish two things which are close in time, just as a higher magnifying power allows you to discriminate between two things close in space.
• Most changes are selectively neutral - Most of the evolutionary changes that we see in the mitochondrial genome do not change the function of the respective genes. These changes do not alter the amino acids encoded by the genes. They are neither advantageous or disadvantageous to the organism. From the point of view of selection they are neutral. As a consequence, these changes accumulate at a relatively constant rate, giving us an evenly ticking molecular clock.
• They are maternally inherited - Each embryo inherits its mitochondria from the maternal egg; the sperm cell does not contribute any. As a result, every individual's mitochondria came from their mother, and hers from her mother, and so on back in time. Genetic material in the nucleus comes from both parents, and meiosis in those parents results in the individual receiving chromosomes from a mixture of ancestors. With mitochondrial genomes the inheritance is simpler, and as you move back through the genealogy you more quickly reach the common ancestor of the genomes sampled.

Every region in our genomes will show a similar type of coalescing back to a common ancestor. A different ancestor is expected for each region in the nuclear genome. Only a single common ancestor is expected for the mitochondrial genome because it does not recombine. The most recent common ancestor of the mitochondrial genome is expected to be much more recent than for other genomic regions because of its pattern of inheritance.