Wednesday, March 12, 2008

Autosomal DNA Testing and Analysis

Unlike yline and mtdna testing where the DNA of the father or mother is passed to the offspring unmixed with that of the other parent, autosomal testing tests all portions of the DNA of an individual. As the field of genetic genealogy has moved forward, research has begun to indicate that certain markers are found in higher or lower amounts in different ethnic populations.

For example, if someone has the Duffy Null allele, or genetic marker, we know they positively have African admixture. We don’t know how much African admixture, or from which line, or when that individual with African admixture entered their family tree, but we know for sure they existed.

Attempting to determine the population frequency of varying markers and what that means relative to other populations is the key to this analysis. Few markers are simply present or absent in populations, but are found in varying frequencies. Some populations are widely studied in the research literature, and others are virtually untouched. The process of compiling this information in a meaningful manner so that it can be analyzed is a formidable task, as the information is often found in nearly inaccessible academic and forensic research publications. It’s difficult to determine sometimes if the DNA analysis of 29 individuals in a small village in northern Italy is, for example, representative of that village as a whole, of northern Italy, or more broadly for all of Italy as a whole. Is it representative of Italy today or Italy historically? These and other similar questions have to be answered fully before the data from autosomal testing can be useful and reliable.

If the DNA tests being performed aren’t mtdna or yline, then they are autosomal tests, meaning they are performed on the balance of the DNA contributed by both parents to an individual.

Before we discuss the varying kinds of autosomal tests and what they mean, let’s take a look at the inheritance process and how it really works.

Inheritance

Everyone knows that you inherit half of your DNA from your mother and half from your father. However, this isn’t exactly true. While each child does on average receive half from each parent, the actual inheritance pattern varies much more than that and each sibling may receive far more than half of their markers from either parent.

We don’t understand today how inheritance traits are selected to be passed to children. Some “groups” of genetic material are inherited together, and you may wind up with more or less genetic material from one of your parents. In time, certain genetic “traits” will be lost in some descendants, while not in others. Therefore, you can’t figure actual inheritance percentages by using the 50% rule. This means that if your father was 50% Native American, you are not necessarily 25%, genetically speaking. You may receive 40% Native genes and your sibling may receive 60%.

Let’s use the Duffy Null allele we mentioned earlier as an example. This marker could have entered your DNA pedigree chart with a grandmother who carried the allele but had no obvious visible African ancestral traits, or from your father who might have been visibly African in ethnicity. The Duffy Null allele, which is just one marker, could have been passed in the inheritance of DNA for many generations, far after any visible African traits had disappeared, or it could be one of many African traits passed from parent to child.

The relevance of the Duffy Null allele is determined by the number of other “African” markers that appear in high quantity. If there are few other African markers, then your African ancestry was likely further back in time. If there are many, then your African ancestry was likely more recent. These statistical calculations are how the importance of autosomal markers are determined and how percentages or estimates of ethnicity are calculated.

Any one allele or marker can be lost permanently in any generation. Each child receives one gene from each parent. In the example below, let’s say that the mother carried genetic markers A and B, and the father C and D, and D is the Duffy Null allele.

Mother Father

Markers A B C D

Child 1 A and C

Child 2 A and D

Child 3 B and C

Child 4 B and D

You can see that half the children received the D marker, but each inheritance event was a random recombination of the markers. It is also possible that none of the children would receive the D marker, or all of them would receive it. Statistically speaking, half will receive the marker, but statistics and individual inheritance are two different things. Random recombination is the reason why siblings who take autosomal tests sometimes show significantly different results.

You can also see how a marker that is very old ancestrally, meaning introduced many many generations ago, could be absent in one entire descendant line and present in another line.

From the above examples, we see that we have two variables that we need to deal with when attempting to use autosomal DNA for genealogy.

First, we need to take into consideration inheritance patterns which we can’t determine retrospectively without testing several descendant lines. So, in essence, we can only deal with, and test, what we personally carry today as our genetic inheritance.

The second variable is determining population frequency for a particular marker and understanding its significance to us through comparative population genetics.

This is why autosomal testing can give us important hints, but are often considered unreliable. The results are highly subjective today, but increase in accuracy as more research is completed, compiled, published and analyzed.

Types of Autosomal Tests

There are two types of autosomal tests used today for genetic genealogy. One type of test uses the Codis forensic markers and the second type, biogeographical tests, use a much broader spectrum of marker results. Let’s look at both types of testing and the information they provide separately.

Codis markers are a standardized set of autosomal markers used for paternity testing. Additionally, they are used by police departments and forensics labs. The markers employed in these tests are selected specifically to differentiate between people in order to identify them individually, not to find common markers to place them in ethnic groups.

The results from these tests are only numbers, and the recipient is often left to their own devices as to how to interpret the results. These tests are available from numerous sources. I prefer to interpret these results in conjunction with Yline and mitochondrial DNA test results for as much of the genetic pedigree chart as can be provided in order to obtain a more complete genetic picture.

Below is an example of what Codis test results look like. They are very similar from any lab.







    Location Mother Child

CODIS example

Analysis of Codis Markers

Unless you?re using the Codis marker results to determine siblingship or some other personal reason, these numbers are fairly useless genealogically. It?s the analysis of these markers that matters.

There are different avenues to analyze Codis results. None are ?right? or ?wrong?. DNAexplain (www.dnaexplain.com) provides analysis of these tests, along with broader more comprehensive analysis of genetic genealogy and what all of these tests together mean about you.

We use a combination of resources, both public and private, including Omnipop and other European and Canadian autosomal forensic data bases.

Tribes (www.dnatribes.com) has been compiling population data on these genetic markers for some years now and will compare your autosomal results with their data base. Take a look at their samples tab.

Ironically, the results may vary significantly between these resources. There is no right or wrong answer at this point. I encourage everyone to simply view these results as data, hints to puzzle pieces. As the data bases improve and we better understand population migration and movement, the clarity of the results will improve too.

Tribes early population tables did not include data from the British Isles, so their results were highly skewed towards other world populations. Omnipop today relies on self-reported ethnicity and does not include normalized data (or a normalizing factor) for varying populations. Because Tribes is a private company, we don?t know much about their population data, whether it's widely representative of the world population distribution and whether it has been normalized or not.

To learn the most about your autosomal test results, you can take a dual approach, having them analyzed by Tribes as well as by DNAexplain using the other autosomal codis reference tools. We?ll be glad to help you through this process and provide a summary analysis of both.

Biogeographical ancestry testing

Biogeographical ancestry testing, available from DNA Print Genomics (www.dnaprint.com), is the second type of autosomal testing. They test all of your genetic contributions for specific, proprietary markers that indicate geographical heritage, not just the Yline or mtdna. They do not use the Codis markers, but use, depending on your test selected, between 500 and 1349 markers they?ve discovered to be relevant to ethnicity.

This test is currently available from only one source, although the test is resold by several testing companies. Results from this test are returned as percentages of ethnic heritage as shown below.


Your results are reported within confidence bands, which indicate a range of percentages that might actually be accurate. This is shown above by the bands surrounding the red dot which shows the ?most likely? result. The margin of error is often as high as 15%. Typically, there is no dispute over the majority ancestral type. However, minority types are apparently much more difficult to discern.

Summary

There are only two tests that can provide you with solid evidence of the source of your Native American or other ethnic ancestry. Those are yline and mitochondrial dna tests. It?s important to try to fill in the blanks in your family tree pedigree chart by testing relatives who carry the yline and/or mtdna of the lines of your tree that you cannot personally be tested for.

In addition, two types of autosomal testing can provide useful clues as to the percentage of your ethnic heritage and the geographical source. Percentages of the 4 major world populations (Native American, African, Indo-European and Asian) are available using the DNA Print test.

Codis marker testing is another type of autosomal test used to determine the Codis marker values which in turn can be used to map those marker values against known population groups. Tribes provides this service for an additional fee using their own internal database.

DNAexplain provides autosomal analysis services for Omnipop and other public databases in addition to analysis services for yline and mtdna test results.

All genetic genealogy results need to be accompanied by genealogical research to unravel the historical context for the lives and trials of our ancestors. DNA testing may well answer the question what and who, but the why is typically revealed only by studying the history of the times in which they lived.

Copyright 2007, Roberta Estes, all rights reserved


http://www.dnaexplain.com/

No comments: