I’ve had various DNA tests in the hope of extending my family tree, but on the whole the result has been disappointing. I have many good autosomal DNA “matches” so I know I am definitely related to those people, but despite their best efforts and my own, we haven’t been able to establish how we are related. For each good match there will be a common ancestor within the last three or four hundred years, but we can’t identify any of them. There’s much more work to do using what’s called phasing and triangulation groups, so I haven’t given up, but it will take a lot of time and effort. (My wife has had more luck and now knows exactly how she is related to one of her matches.)
To do these searches etc there are various websites offering tools, mostly free, to automate the work, and some offer “fun” DNA tools too. One company, GEDmatch, offers a tool which compares anyone’s DNA with DNA extracted from the bones of ancient people found in various archaeological digs around the world. This tells me that I’m a close match to the DNA of persons who died about 7000 years ago at Stuttgart, Germany; 7200 years ago at Polgár-Ferenci-hát, Hungary; 8000 years ago at Loschbour, Luxembourg and 45000 year ago at Ust’-Ishim, Siberia; and a frighteningly close match to the DNA of a person, a man, who died 3200 years ago (Late Bronze Age) at Ludas-Varjú-dűlő, Hungary. Fascinating stuff, but not really of genealogical interest.
The “fun” tool which interested me particularly is another from GEDmatch called “Are your parents related?” As its name suggests, it looks at my raw DNA data and tells me to a good degree of certainty whether or not my parents were related to each other. Everyone is related to everyone else, of course, but here I’m talking about a comparatively close relationship with a common ancestor within the last six or seven generations.
The principle is simple. Most people will be aware that our DNA consists of a double helix divided up into 23 sections of varying lengths called chromosomes. Less well known is the fact that one strand of that double helix came from our fathers and the other from our mothers. They, of course, each got their DNA from their ancestors and this will be distributed in blocks, called segments, more or less at random along their DNA strand – some ancestors contributing a lot, others little, and, perhaps surprisingly, some ancestors contributing nothing at all. In general, recent ancestors will have contributed more segments and larger segments, distant ancestors will have contributed fewer and smaller segments.
The two strands of my DNA double-helix – one from my mother and one from my father – will, of course, have a huge number of differences over most of their lengths, but if I compare the two strands using this GEDmatch tool and find significant segments which are identical on both strands, we can say with certainty that each shared segment of DNA originated from one person in the past and has been passed down generation by generation separately to each of my parents. In other words, if that were the case, my mother and father must have had a common ancestor and were therefore related.
When I ran the tool on my DNA data I really wasn’t expecting anything to show up. None of my genealogical research so far has suggested that my parents were related. As I slowly scrolled down through the results chromosome by chromosome, this seemed to be confirmed as there was nothing showing up. I scrolled faster , then suddenly – whoa, what was that? A matching segment on chromosome 17, and quite a big one at that.
Matching segments are measured in centimorgans (cM, not to be confused with centimetres cm) and Single Nucleotide Polymorphisms (SNPs). Any matching segment above 7 cM and 700 SNPs is considered potentially “interesting” and the greater the two measurements above those figures, the more certain one can be that the match is “genuine” and not down to down to coincidence. The match on my chromosome 17 is of 17.0 cM and 2282 SNPs, which is well above both thresholds and almost certainly represents a real match – the segment really must have been passed down to me through both sides of the family from a common ancestor.
|Chr||Start Location||End Location||Centimorgans (cM)||SNPs|
|Estimated number of generations to MRCA = 4.9|
So my parents were definitely related, and the GEDmatch tool calculates that the most likely number of generations to MRCA (Most Recent Common Ancestor) is 4.9 – say 5. Of course this is just the most likely – it could be slightly fewer or slightly more than 5 generations, and there’s a very small chance that it could be a lot more than 5.
Unfortunately I don’t yet know who that MRCA is. Of course, I don’t know all my ancestors even 5 generations back from my parents. In fact I know the identities of only 27 of my dad’s 32 in that generation, and only 20 out of the 32 on my mum’s side, and I know an even smaller proportion in earlier generations. The MRCA must either be an ancestor I haven’t yet identified at all, or one I have only identified on one side of the family; a really good incentive to double my efforts to extend my family tree backwards in as many lines as possible.
The only clue I have at present points to the Allendale area in SW Northumberland as a potential location for this MRCA. That was the early home of the Pears and Philipson lines on my dad’s side, and of the Henderson line on my mum’s side. It’s the only place I’ve found so far with connections to both sides of my family and it’s certainly where I’ll concentrate my researches first. As Sherlock put it: “The game is afoot.”