What is BioGeographical Ancestry (BGA)?
BioGeographical Ancestry (BGA) is the term given to the biological or genetic component of race. BGA is a simple and objective description of the Ancestral origins of a person, in terms of the major population groups. (European, East Asian, sub-Saharan African, Native American, etc.) BGA estimates are able to represent the mixed nature of many people and populations today. In the US, as in many other countries across the globe including the UK, there has been extensive mixing among populations that had initially been separate. In the fields of human genetics and anthropology, this mixing is referred to as admixture. BGA estimates can also be understood as individual admixture proportions, which take the form of a series of percentages that add to 100%. For example, a person in question from the United States may be found to have: 75% European ancestry; 15% African ancestry; 10% Native American ancestry, or they may be found to have 100% European ancestry.
What is the Relationship between Race and Genetics?
First, it is important to realize there are far more genetic similarities between all human beings than there are differences, regardless of the racial heritage. At the level of DNA, only one base in a thousand differs between the genomes of any two people. The term race, in general usage, includes both a cultural and biological feature of a person or group of people. Given the fact that physical differences between populations are often accompanied by cultural differences, it has been difficult to separate these two elements of race. Over the past few decades there has been a movement in several fields of science to oversimplify the issue declaring that race is "merely a social construct". While, indeed this may often be true, depending on what aspect of variation between people one is considering, it is also true that there are biological differences between the populations of the world. One clear example of a biological difference is skin color. There is a strong genetic component to the level of pigmentation in a person's skin and there are dramatic differences across populations. Pigmentation is literally only skin deep, a heritable trait under natural selection across the complex environments in which we all live. For example pigmentation is driven by the contrasting need to protect our skin from UV-light damage while allowing sufficient exposure to sunlight to catalyze the light-driven conversion of vitamin D precursors into the active form of this essential vitamin. Many other examples of genetic selection in human populations are under active scientific investigation.
The biological component of race is largely based on the genetic structure of human populations. This structure is a nested hierarchy from East to West where populations in the Americas and the South Pacific are a subset of the genetic diversity found in Eurasia which itself is a subset of the diversity found in Africa.
It is clear that the human species is relatively young. As a species, we most likely originated in east Africa, according to most archeologists, 100,000 to 300,000 years ago, and diverged as groups, expanded, moved, and settled the globe. During these migrations, and in the time since, there has been some degree of independent evolution of the populations that settled the various continents of the world. The simplest evidence of this evolution can be seen in the differences in the frequencies of different versions (“alleles”) of genes. Generally, we see that alleles found in one population are also found in all populations and the alleles that are the most common in one are also common in others. These similarities between populations highlight the recent common origin of all populations and strong connections between populations throughout human history. However, there are examples of DNA variations, also called genetic markers, which can be used to estimate the ancestral origins of a person or population.
It should be clear by now that race is a phenomenon dependent upon multiple variables. In our procedures we focus on a person’s genetic ancestry and not their race. Your DNA has no recorded history of your political, social, personal or religious beliefs. It is a simple code written in four letters that records all of the changes in the DNA from one generation to the next. We identify and report those variations within the DNA code that reflect your genetic ancestry most succinctly. They are like fingerprints and snow flakes, unique and wildly complex.
How is BioGeographical Ancestry estimated?
The AncestryByDNA test uses an especially selected panel of Ancestry Informative Markers (AIMs) that have been characterized in a large number of well-defined population samples. These markers are selected on the bases of showing substantial differences in frequency between population groups and, as such, can tell us about the origins of a particular person whose ancestry is unknown. For example, the Duffy Null allele (FY*0) is very common (approaching fixation or an allele frequency of 100%) in all sub-Saharan African populations. Thus, a person with this allele is very likely to have some level of African ancestry. After the analysis of these AIMs, in a sample of a person's DNA, the likelihood (or probability) that a person is derived from any of the parental populations and any of the possible mixes of parental populations is calculated. The population (or combination of populations) where the probability is the highest is then taken to be the best estimate of the ancestral proportions of the person. Confidence intervals on these point estimates of ancestral proportions are also being calculated.
How can BGA estimates be used?
We believe that an objective assessment of the biological component of human ancestry is possible and that such research could enrich our lives in a number of ways.
Estimates of BGA can help reconnect individuals separated by adoption, or some other event, with their ancestral populations. Even if a person is not particularly motivated to reconnect with ancestors, he or she can uncover the past of their family either to verify family legends or to search for forgotten roots.
In the near future, we hope to allow customers to compare their ancestral proportions to others in their family, town, city, or region who have chosen to participate. Because it is based on DNA this new tool will provide the most accurate demographics data possible.
What is the medical significance of BGA estimates?
The medical significance of BGA estimate is negligible at this point in time. Although some diseases are found at different frequencies in populations across the globe, hardly any are restricted to one group. The usefulness of BGA estimates, in biomedical research, comes from epidemiological analyses where many individuals are analysed together to make general statements about differences in risk. Even though these results can be significant, they are not foolproof predictors with regards the level of risk for any one person in the population.
How is BGA analysis different from mitochondrial DNA and Y-chromosomal ancestry analysis?
There are several commercially available tests of mitochondrial DNA (female contribution) and Y-chromosomal DNA (male contribution) markers, which have been promoted as a means of learning one’s ancestral origins. Although these tests could provide information regarding the provenance of some of a person’s ancestors, limited in their diagnostic power. For example, one generation ago a person has two ancestors, one mother and one father; five generations ago, a person has 32 ancestors; while 10 generations ago, a person has 1024 ancestors. Ten generations is roughly 250 years and within the time frame of genealogical interest, especially when we are considering the settlement of North America, because they only look at two (2) chromosomes. Y-chromosomal analysis and mitochondrial DNA analysis can each could only provide information on a very small proportion of a person’s ancestors. Our test relies on sequences throughout your genome, so we can say more about a greater number of your ancestors
BGA provide more specific information about ancestry?
AncestryByDNA 1.0 is the first version of the test, and it is specifically designed to provide information on the proportions on ancestry on the continental level. In other words, this test allowed us to uncover the levels of European, African and Native American as three component groups. The current BGA tests that is offered, AncestryByDNA 2.0 and 2.5, are expanded to provide information on the proportions of ancestry on the continental level for most continents, European (which includes European, Middle Eastern and South Asian groups such as Indians), African, Native American, but we distinguish ancestries within Asia and the Pacific Rim by adding East Asian, (which includes the Pacific Islanders) as an additional group. Since there will ultimately be interest in defining the levels of ancestry within continents (such as distinguishing Japanese from Chinese, or Northern European from Middle Eastern), we are in the process of developing a new series of Ancestry Informative Markers that will provide more insight into where within a particular continent a persons’ ancestors were most likely derived. This test, AncestryByDNA 3.0 is currently under development.
Have there been any recent changes to the test?
We have improved the software for generating and presenting results. The current version of the software builds on the knowledge gained during the development of AncestryByDNA 2.5. Prior to the change, we reported the results in terms of a single triangle plot, showing the confidence contours in the three dimensions corresponding to the three most likely ancestries for the individual. The new software shows BGA estimates using a more complex triangle plot and a bar graph. The new triangle plot and bar graph show the confidence contours in all four dimensions (European, Western Sub-Saharan African, European, East Asian and Native American).
My proportions were 85% European and 15% East Asian, and the East Asian part surprised me. How reliable is the 15% part of the profile?
The 15% East Asian means you most likely share sequence identity with East Asians at some markers. Studies have indicated that, although there is a measurable level of "noise" in the test, we do see a pattern of minor contribution among certain populations. These findings will be published at a later date.
Your range bars and confidence contours given with the analysis are also part of your answer. The values within these determined ranges represent other possible outcomes that are statistically significant but are less likely. Therefore, if your range bar on the bar graph includes zero, you should consider that possibility.
How can I confirm the significance of a low percentage of admixture?
There are two ways for you to confirm the value of this estimate:
You may have access to historical records that leads you to confirm or refute the admixture. For example, if your records suggest that you have a grandparent of East Asian heritage and you register with the test as of 5% East Asian, the two observations combined make a stronger case for East Asian ancestry than either on their own.
You can obtain the admixture proportions for your father and mother. Lets say you register with 4% African and you want to know whether this 4% is in error or is accurate. You obtain the admixture proportions from your parents and each is 100% European. Chances are the 4% was a result of statistical noise. However, if your mother was 15% African and your father was 100% European, your non-zero percentage of African is likely to be an accurate indicator of African ancestry.
I thought I was purely of Scandinavian origin, but my results show minor East Asian admixture. How is this result possible?
In our testing we’ve observed a number of samples from Scandinavia and Eastern Europe that exhibit some East Asian admixture. This is probably due to past migrations of Nordic groups into Asia and is possibly telling us something about the interaction of various groups in those regions. In addition, cultures from these areas traveled to region of Asia and it does not seem unreasonable that children may have been produced from some of these travels thereby introducing new genetic markers into the Scandinavian population.
How accurate are admixture scores of less than 10%?
Pedigree studies have shown that the levels of admixture follow known inheritance patterns, supporting the validity of these low values. However, as stated above, the confidence intervals are important for understanding your results and the scores are the most likely estimates from your DNA. If you confidence interval overlaps zero your value may be within the statistical noise threshold of the test and should be considered a possible result. As the number of markers tested goes up the statistical noise should decrease as evidence in the change in confidence interval size between the AncestryByDNA 2.0 and AncestryByDNA 2.5 tests.
Has this technology been published in the scientific literature?
Yes, here is a selection of relevant publications:
Parra, E., Marcini, A., Akey, J., Martinson, J., Batzer, M., Cooper, R., Forrester, T., Allison, D., Deka, R., Ferrell, R. and M. Shriver (1998). Estimating African American Admixture Proportions by Use of Population Specific Alleles. American Journal of Human Genetics. Volume 63, pages 1839-1851. Pfaff, C., Parra, E., Bonilla, C., Hiester, K., McKeigue, P., Kamboh, M., Hutchinson, R., Ferrell, R., Boerwinkle, E., and M. Shriver (2001). Population Structure in Admixed Populations: Effect of Admixture Dynamics on the Pattern of Linkage Disequilibrium. American Journal of Human Genetics, volume 68, pages 198-207.
Parra, E., Kittles, R., Argyropoulos, G., Pfaff, C., Hiester, K., Bonilla, C., Sylvester, N., Parrish-Gause, C., Garvey, W., Jin, L., McKeigue, P., Kamboh, M., Ferrell, R., Pollitzer, W., and M. Shriver (2001). Ancestral Proportions and Admixture Dynamics in Geographically Defined African Americans Living in South Carolina. American Journal of Physical Anthropology 114:18-29.
Frudakis, T., V Kondragunta, M Thomas, Z Gaskin, S Ginjupalli, S Gunturi, V Ponnuswamy, S Natarajan, and P Nachimuthu (2003). A Classifier for SNP-Based Racial Inference. Journal of Forensics Sciences, Volume 49, pages 1145-1146.