This is the first part in the third and final ‘mini-series’ in my beginners’ guide to genetic genealogy. You’ll find links to all the previous posts [here].
In this mini-series we’re moving on to something called ‘chromosome mapping’ and as an essential part of that we’ll be looking at my favourite tool: the chromosome browser. We’ll focus on:
- a definition of chromosome mapping
- what a chromosome browser is
- what it tells you about your match with another person
- how we can use it to identify multiple matches who are descended from the same common ancestors
- amount of shared DNA and length of segments as a guide to the closeness or remoteness of a match
- a case study illustrating how this can all work together
What do we mean by chromosome mapping?
When you have a number of confirmed DNA matches and you’ve identified Most Recent Common Ancestors (MRCA) for each of them, you have already started to allocate matches to different branches of your tree. A second cousin, for example, becomes a ‘benchmark’ for more distant DNA matches along the same great grandparent line. We can do this far more accurately if we can actually see the specific segments we share with that second cousin. All these segments, we know, came to us (and to this second cousin) from that known set of great grandparents. If we now find a more distant match (say, a 4th cousin) on one or more of the same segments we can trace those specific segments back even further. Not only will this help to verify our documentary research going back to the next MRCA – likely to be around 3xG grandparents – but also we now know *which* of those previously identified great grandparents this specific segment of our DNA actually came from. I find that very exciting. However, before we can do any of this we need to be able to ‘see’ those segments, and for that we need a chromosome browser.
What is a chromosome browser?
Essentially, a chromosome browser is a visual representation of the parts of your DNA that you share with one or more other people. It looks like a series of ‘stripes’ – one stripe for each of your twenty-two autosomal chromosomes. If you tested with FTDNA or 23andMe there will be an additional stripe for the X chromosome.
The following is an example of one of my matches from MyHeritage. I have removed name and contact details, but we’ll call this person A.
The essential information provided here about my match with A on the general listing of all my matches is:
- we share 155.4cM across eight segments
- this equates to 2.2% of our autosomal DNA
- the longest segment is 45.5cM
- our estimated connection is ‘1st cousin twice removed – 2nd cousin once removed’
If I click on the pink ‘Review DNA Match’ button, I also see
- a list of all the other people on the MyHeritage site who also match A and me
- if A has a tree on MyHeritage I can look at that
- lists of any surnames we have in common
- a comparison of our ethnicities estimates
- a chromosome browser representing our match
That’s a lot of information.
At its most basic level, the chromosome browser provides a graphical view of some of that information. It doesn’t provide any surname or ethnicity data, but it does show exactly where, on your chromosomes, those shared segments are.
Below is that match with A as viewed in the chromosome browser. The 22 grey lines represent me: my 22 autosomal chromosomes. The segments where A and I match are those pink lines (segments) on chromosomes 1, 7, 11, 13, 17, 19 and 21. Although you can’t see it on this screen grab image, if I hover my cursor over any of the pink segments while I’m on the MyHeritage website I can see exact locational information (start and end points). I can also download all that information.
These segments are where A and I have inherited exactly the same DNA from the same ancestors.
Using the chromosome browser ‘One to Many’ function
A and I now have a lot of information about our match but although we know we’re fairly closely related we still can’t say which of our ancestors we both share. For that we need to look at our documented trees. However, in this case, by a process of elimination of A’s close family members who have tested but don’t match me we have been able to conclude that we are connected on the one part of A’s tree about which nothing is known: a missing grandparent. The chromosome browser may be able to help.
Using the ‘One to Many’ tool, we can use the chromosome browser to compare overlaps between our own DNA and that of several other people – up to seven matches all in one go at MyHeritage. In the example below I’m looking at just two of my matches, comparing exactly where each of them matches me, and looking for any segments where all three of us match. Again, the grey lines represent my 22 autosomal chromosomes. The red segments show my match with A (you can see they are in exactly the same positions as in the above chart, although now they are coloured red). The mustard segments are another person with whom we both match at a close level. We’ll call that person B.
The chromosome browser here shows that I share more DNA with B (mustard) than I do with A (red), and the segments tend to be longer. It also shows that I share lots of DNA with each of them that I don’t share with the other. However, there is one more very important piece of information: A, B and I all share exactly the same DNA in two positions: on chromosomes 7 and 13. Taken as a whole this tells us:
- All three of us are descended from the same fairly recent common ancestors.
- Between the three of us, we have inherited different parts of the DNA of these common ancestors. What we inherited has come down the line from them to us via our own parents and grandparents – and just to recap a key point from my post earlier this year (Asking other family members to test) siblings (in this instance our respective grandparents) inherit a lot of the same DNA but not exactly the same – which explains why A, B and I don’t all have exactly the same autosomal DNA from those ancestors
- I am more closely related to B than I am to A
- Since I have previously placed B as my second cousin I already know which set of great grandparents are our Most Recent Common Ancestors (MRCA). Those two 3-way shared segments are proof that A is also descended from these same common ancestors. As a result A now has a name for that missing grandparent
- Also as a result, I now know that all those segments coloured red and mustard on my maternal line have come to me from that particular set of great grandparents
What we have just seen at chromsomes 7 and 13 is an example of triangulation. To fully understand what this means, we need to understand a key point about chromosome browsers. The chromosome browser represents each chromosome as a single grey line, but that one grey line belies that fact that we get two copies of each chromosome: one from our mother and one from our father. The chromosome browser cannot tell these maternal and paternal lines apart. Therefore when you look at a match on a chromosome browser the segments on those grey lines could be maternal or paternal. They could even (more unusually) be a mix of both if your match is related to you via your maternal and your paternal lines. We have to find some way of working out which.
If you’re working on MyHeritage the Triangulation tool helps with this. It tells you if two or more people match with you at a given segment on the same copy of the same chromosome. In other words – either on your maternal side or on your paternal side. It still doesn’t tell you which side that is, but if you can work it out for at least one of the matches using your documented tree, then you know that this is also where the other ‘triangulated’ person matches you.
You can see how the triangulation tool works in the example above: there is a little box around the overlapping segments at chromosomes 7 and 13. This is MyHeritage’s way of saying that these two segments of DNA shared by A, B and me are indeed on the same copy, and we are therefore descended from the same line. A, B and I triangulate.
Segment length and amount of shared DNA as a guide to remoteness of the match
Obviously, these great grandparents from whom A, B and I inherited all these DNA segments, they too inherited their DNA from their own parents, and those parents from their parents, and back through time. Since we all inherit 50% of our autosomal DNA from our mother and the other 50% from our father, when my mother was born she recombined all of this DNA from the line we share with A and B with a whole new set of DNA from her other parent, and then I did the same. This means that DNA shared with closer matches (and inherited from closer ancestors) tends to have longer segments: there have been fewer recombinations. By contrast, shared segments with more distant matches have undergone more recombinations – the DNA is mixed up with that of far more other ancestors – and so segments tend to be shorter. Hence, alongside the total amount of shared DNA, we can also use segment length as a guide to closeness or remoteness of a match.
Once we have identified the MRCA (or MRCA couple) from whom we inherited a decent amount of autosomal DNA (as I did above with A and B) we can use that information to try to place other people who match us more remotely on any of the same segments.
Consequently, attributing a segment to one ancestral couple is not the end of the story because they got it from someone else – or rather one of them did. Therefore when we use new, more distant, DNA cousin matches to take that segment (or part of it) back further, we can work out which of the closer ancestral couple it came from, and which of their ancestors they (and we!) got it from.
A more remote example – a work in progress!
The final image, below, shows another One-to-Many chromosome browser example. This time the red lines are B (who was mustard in the last example) and the new person (mustard segments) is C. You can see here that I share far fewer segments with C, and that the segments are shorter. We actually share 31.9cM, and the longest segment is 17.4cM. Using all the information outlined above, we can say that my match with C is more remote than my match with B. Our estimated relationship is given as ‘3rd to 5th cousin’, suggesting MRCAs at 2xG grandparent to 4xG grandparent. Importantly though, MyHeritage’s triangulation tool (the box on chromosome 11) tells me that this match with C is further back along that same line that I share with B. If I can find where C and I connect I will know exactly which more distant ancestors these mustard-coloured segments of my DNA come from. But of course, beyond that set of great grandparents that I share with B there are four GG grandparents, eight 3xG grandparents, sixteen 4xG grandparents, and so on. Our MRCA could be any one of them.
So this one is still a work in progress. By making contact with C, I was able to use the same triangulation process (with one of his closer cousins) to determine which of C’s great grandparent lines we match on. So we know which of *his* great grandparents and also which of *my* great grandparents, but so far analysis and further building of his tree has drawn a blank.
I have reason to suspect that these short segments have come to me from an Irish 3xG grandmother along this line, precise origins unknown, but this is not my common ancestor with C. If I can find an Irish connection along the relevant part of C’s tree, or indeed if I can find a definite Irish lineage for one or more additional people who triangulate with B, C and me along this line, then I may even be able to place this Irish 3xG grandmother’s origins in a specific part of Ireland. I have some ideas for how to progress this, and if they work I’ll write about them in a future post.
In my next post I’ll be continuing this theme by looking at DNA Painter, a third-party tool that helps with the organisation of segment information.
My DNA posts are intended as a beginners’ guide, building up the information in order, in bite-sized chunks. Click [here] to see them all in the order of publication.