Skip to main content

A Little About Autosomal DNA Testing

Image
By

This post is an adaptation of a post I originally published as a blog post at Ancestorian. The original blog post was written in response to an Ancestorian member's question. The member asked:

We are having our DNA tests now, and showing up against living family members we know … HOW is that DNA test proven against ancestors 200 years ago against their DNA which we don’t have.

The inquiry related to autosomal testing (At-DNA). The inquirer asked for a simple (that is, basic) answer to her question. This post, more or less, details the answer I gave; yes a little long-winded but I felt it necessarily to underpin the answer with a little background information about autosomal DNA testing.

Autosomes

Autosomal testing (At-DNA), as its name suggests, looks at what can be learned from the data you carry on your autosomal chromosomes. You have 22 pairs of autosomes. That is, one copy of each of the autosomal chromosomes 1-22 from each parent. You also have 1 pair of sex chromosomes. That is one copy from each parent. Males are xy. They get the y copy from their father and the x copy from their mother. Females are xx. They get one x from each parent. The x chromosome is sometimes referred to as chromosome 23.

While some of the companies include x chromosome data, others such as Ancestry focus only on the 22 pairs of autosomal chromosomes when determining DNA matching and shared cM values.

For the most part, the pattern of DNA inheritance for the sex chromosomes differs to that of the 22 autosomes. It requires separate explanation. My focus in this post is on the 22 autosomal chromosomes.

DNA Recombines Randomly

DNA is mixed up at each generational level in a process called recombination. Recombination is the exchange of DNA between the two copies (one maternally inherited and one paternally inherited) of each chromosome (1-22).

During the creation of an egg or sperm for the next generation, DNA segments are swapped between the two copies of each autosome at one or two random spots on each of the autosomes (crossover points). The newly formed chromosome in the egg will be a mix of what the maternal grandparents have contributed and the newly formed chromosome in the sperm will be a mix of what the paternal grandparents have contributed.

So:

  • The way in which DNA recombines at each generation is random; and
  • Which bits of your ancestors’ DNA you get is random

This also means that, because the creation of each egg and sperm is an independent event, siblings (identical twins aside) don’t inherit exactly the same portions of their grandparent’s DNA as one another. They share some in common but not all. 

In terms of your 22 pairs of autosomes, you inherit 50% of the DNA from each of your parents. On average, you inherit 25% of your DNA from each of your grandparents (via each of your parents), 12.5% from your great grandparents, 6.25% from 2nd Great Grandparents, 3.125% from 3rd Great Grandparents … and so on. In reality, it is likely you will get slightly differing amounts of DNA than the average amount referred to above from each of your grandparents. The variance from average is likely to increase as you go back generations beyond your grandparents. This is because of the random way in which DNA passes from one generation to the next. 

As indicated above, 25% is the amount of DNA you inherit, on average, from each of your grandparents. If you are a little short for one grandparent within an ancestral couple, you will receive more from the other grandparent within the ancestral couple but the amount will average out across the two grandparents within the ancestral couple. Similarly, your parents may have received slightly differing amounts from each of their grandparents and your grandparents may have received slightly differing amounts from each of their grandparents, and so on. So it is that DNA washes out unevenly over time (See below).

Useful links:

Your Genetic Tree is a Subset of Your Genealogical Tree

The number of ancestors you have doubles with each generation:

Number of Ancestors by Generation

When you consider how the number of generations increases moving back in time and the random way in which DNA recombines at each generation in order to move forward in time, it is not difficult to understand how the percentage of autosomal DNA coming from each ancestor is diluted with each new generation. Because segments from specific ancestors are passed down randomly, they can be quickly lost.

Because DNA washes out over time, you will have the DNA of some of your ancestors only. Somewhere between about 6-8 generations back, or so, ancestors start falling off your genetic tree. 

The following table from ISOGG Cousin Statistics - https://isogg.org/wiki/Cousin_statistics - shows the probability of sharing no detectable DNA relationship with your ancestors:

https://isogg.org/wiki/Cousin_statistics

Also, because DNA is passed randomly, you will have relatives more distant than 2C who belong in your genealogical family tree but who do not share a detectable amount of DNA with you because they got different bits of DNA from your common ancestors than you did or, for more distant relatives, they inherited DNA from an ancestor when you didn’t. They are genealogically related to you but not genetically connected to you through shared DNA.

The following table from ISOGG Cousin Statistics - https://isogg.org/wiki/Cousin_statistics - shows the probability of sharing no detectable DNA relationship with your relatives: 

https://isogg.org/wiki/Cousin_statistics

You can only be matched to other people who share a detectable amount of the same bits of DNA with you. Because you don't inherit exactly the same DNA as your siblings, your DNA match list will look a little different to theirs for matches beyond second cousins (2C). It is quite possible, therefore that a sibling will match say a third cousin (3C), for example, when you don’t.

Useful links:

Making Connections

Where the testing company finds that you share a detectable amount of DNA of a reasonable size with another person in their database you are matched to that person. The company also tells you which other matches match you both. Don’t under-estimate the value of shared matching. It is a really useful tool for sorting your matches into groups or clusters.

Some testing companies tell you where on a chromosome you share DNA with a match. Where 3 or more matches (on the same side of the chromosome - that is, already identified as paternal or maternal) match each other in the same place on the same chromosome, this indicates that they share a common ancestor. That is, that the same chromosomal segment has been passed down to you and each of the shared matches from the same ancestor.

Ancestry doesn’t provide detailed chromosomal information but, as with the other databases, you can make good use of shared matching to figure out groups of matches that all match one another. If you can’t connect them to your own tree straight away, you can investigate how they link together around a common ancestor. This may be a clue to one of your brick walls. 

Matches that group together in a cluster form an 'In-common With' (ICW) group. This is not the same as a Triangulation Group (TG).  ICW matches will connect to one another somehow but the entire ICW group will not necessarily triangulate around a common ancestor (but sometimes do). 

Keep an Open Mind

It is important to keep an open mind as you triangulate the information you have available from DNA testing and documentary sources and work to connect your matches together around a common ancestor. Figuring out which line the DNA has passed down to get to you isn’t necessarily as simple as connecting the shared matches to the same ancestor in your tree. Genealogically, you may connect to a match on multiple lines but the genetic connection may only be from one of those lines. Depending how complete your tree is (see below), you may not be aware of the other line. Sometimes people share more than one segment with a match. Each segment may belong to the same line as the other segment(s), or the segments you share with the match may have been passed down to you from different lines.  

Reliability

At this point perhaps we should return to the question initially posed by the inquirer:

HOW is that DNA test proven against ancestors 200 years ago against their DNA which we don't have?

The short answer is, it doesn't if you don't have their DNA ... not directly anyway.

If we assume that the average span between one generation and the next is about 25 to 30 years, then 200 years could take us back, 8 or more generations. As discussed above, it is quite possible that you won't carry the DNA of some of those ancestors. In that case, your autosomal DNA test won't directly link you with those ancestors. Indirectly, it may, if you have a close relative (e.g., sibling, parent … ) who matches others who link back to the ancestors who you don't share DNA with.

Even then, that in itself is not absolute proof of anything! DNA does not stand alone. You also need to have confidence in the documentary evidence that supports the way in which you and your matches connect together around the common ancestor or ancestral couple. Also, as mentioned above, you need to consider the possibility of more than one connection to the match. Can you be sure that is the line the DNA has come down to your family from? Could there be an alternative line? 

Sometimes people are too quick to conclude that they have proven something even though it is a few generations back and they have gaps in their tree. Questions you need to ask yourself are ... Can I be sure that this is a genetic connection rather than just a genealogical one? How complete is my tree? Could I connect to these matches in some other way as well? 

Tree Completeness

A great way to check how complete your tree is, and to remind yourself where your brick walls are, is to use the ‘Tree Completeness’ function within the ‘Ancestral Trees’ tool at DNAPainter - https://dnapainter.com/#trees.

This is what DNAPainter shows me about my tree completeness:

Screenshot of output from https://dnapainter.com/#trees

You can see that I have some significant gaps as early as 3rd Great Grandparent level. In my case, my paternal ancestry goes back to Colonial America. Endogamy[1] is a feature of early Colonial American populations. As such, it is very likely that I connect to some matches, particularly the more distant ones, on more than one line.

To Conclude

DNA-related information provides additional evidence. It is used alongside documentary evidence etc. to help us hypothesise, or reach conclusions, about our ancestral and genetic connections. It has helped me enormously in places but I have varying degrees of confidence in the conclusions I have drawn and, because I have significant gaps in my tree, my conclusions stand only until other evidence comes along to make me rethink it. For this reason, you won't find the words DNA verified or DNA proved anywhere in my tree. There may well be other people who have a greater degree of confidence in the extent to which their DNA information, alongside their documentary evidence, proves a relationship in their tree based on a more complete and well-evidenced tree.

Notes

[1] Endogamy arises as a result of intermarriage, over a number of generations, within closed or isolated communities where marriage outside the tight-knit community is either highly discouraged or impossible. Closed communities include those where custom or law forbid marriage outside the group for social and/or religious reasons. Isolated communities are those where circumstances (unavailability of marriage partners outside the group) lead to people intermarrying within their own community. For more information about endogamy and the difference between endogamy and pedigree collapse see https://www.yourdnaguide.com/ydgblog/pedigree-collapse-v-endogamy and https://dna-explained.com/2021/07/23/whats-the-difference-between-pedigree-collapse-and-endogamy/

Labels:

Comments

Post a Comment