Newsletter
- 10th July 2016
DNA Special
You've
got your DNA results - what next?
How far
back is our common ancestor?
Why
place names are more important than surnames
How
common matches can lead to common ancestors
Why
you should ask your cousins to test...
... and why you should ask your cousins if they have already
tested!
Time
to update your My Details page NEW FEATURE
More
DNA tips & techniques to come
Mother
wins right to use dead daughter's eggs
A new way
to search for war deaths
How
Newfoundland soldiers forged lasting link with Scotland
Britain's oldest
person dies, aged 113
New
CEO takes over at Findmypast
Try
out Findmypast with a half-price World subscription ENDS FRIDAY
How has
Brexit affected genealogists?
The LostCousins newsletter is usually
published fortnightly. To access the previous newsletter (dated 24th June) click here; to find earlier articles use the
customised Google search below (it only searches these newsletters, so you
won't get spurious results):
Whenever possible links are included to the websites or articles
mentioned in the newsletter (they are highlighted in blue or purple and underlined, so you can't miss them). If one of
the links doesn't work this normally indicates that you're using adblocking software - you need to make the LostCousins site
an exception (or else use a different browser, such as Chrome).
To go to the main LostCousins website
click the logo at the top of this newsletter. If you're not already a member,
do join - it's FREE, and you'll get an email to alert you whenever there's a
new edition of this newsletter available!
You've got your DNA
results - what next?
The best thing about taking an autosomal
DNA test is that you'll get hundreds, perhaps thousands, of matches with cousins
- but how on earth do you figure out how you're related to all of them? I know
from the emails I've had that for most researchers it's quite a conundrum, so
I'm going to tell you what my strategy is....
But first, a reminder that anyone who
takes an atDNA test thinking that it's easier than
doing conventional research is likely to be sorely disappointed. Believe me,
finding 'lost cousins' by completing your My
Ancestors page is a LOT easier, as well as being infinitely cheaper. The
time to turn to DNA is when you've gone as far as you reasonably can using paper records.
If you've just taken a DNA test and are
waiting for the results you've probably got in mind some particular 'brick
walls' that you'd like to knock down. At the last count I had about 70 'brick
walls' in my family tree, but it tends to be the more recent ones - in the late
1700s or early 1800s - that are most frustrating, because they're the ones that
have been outstanding the longest.
Let's suppose that, like me, you've got
a great-great-grandmother who suddenly materialises in London on her wedding
day. There are no confirmed sightings of her before that, and despite following
every clue her birth and baptism simply can't be found.
I have 16 great-great-grandparents, so
each of them is the gateway to 1/16th of my family tree. On average I've
inherited 1/16th of my autosomal DNA from each of them, and - again on average
- about 1/16th of all my DNA matches will be with cousins from that part of my
tree. Since Family Tree DNA currently shows that I have 837 matches, that's around 50 for each of my great-great-grandparents.
Great - I've got 50 chances to knock
down that annoying 'brick wall'! But wait a minute, how do I know which of
those 837 matches are the ones I should be looking at? Indeed how can I know how any of my DNA cousins are actually
related to me?
There's one obvious way to work out how
you're related to your DNA cousins, and that's to look for familiar surnames in
their tree. The first problem with this strategy is that not everyone uploads
their tree, so you would have to email the other person; the second problem is
that in most cases the surname of your common ancestor won't appear in both
trees - indeed, it might not appear in either tree!
In conventional paper-based research we
tend to spend more time working on males lines than female lines for one simple
reason - males generally keep their surname when they marry, females usually
don't. This makes it so much easier to research male lines, and as a result we
don't discover a true cross-section of living cousins, or of deceased
ancestors. However, unlike Y-DNA (which tracks the male line), autosomal DNA is
passed on by both mother and father, which means that your DNA cousins are a
much more representative sample of your living cousins.
The further back the common ancestor,
the less likely it is that you'll recognise any
of the surnames in your cousin's tree, or that they'll recognise any of the
surnames in yours. Surnames will rarely tell you which ancestral line you share
with a DNA cousin for the simple reason that most of your DNA matches are many
generations back, as I'll explain the next article.
How far back is our common
ancestor?
One of the most useful pages on the
International Society of Genetic Genealogy (ISOGG) website is headed up Cousin statistics
- you'll find it here.
The first table shows the chance that
two cousins will be matched by their DNA:
Relationship |
23andMe |
||
First cousins |
100% |
100% |
100% |
Second cousins |
100% |
100% |
>99% |
Third cousins |
89.7% |
98% |
>90% |
Fourth cousins |
45.9% |
71% |
>50% |
Fifth cousins |
14.9% |
32% |
>10% |
Sixth cousins |
4.1% |
11% |
Remote (typically less than 2%)[2] |
Seventh cousins |
1.1% |
3.2% |
|
Eighth cousins |
0.24% |
0.91% |
|
Ninth cousins |
0.06% |
||
Tenth cousins |
0.002% |
|
Looking at this table it's all too easy
to jump to the conclusion that most of your DNA cousins will be close cousins -
but you'd be wrong, because you have many more distant cousins than you do
close cousins.
Here's another table used on the ISOGG
page, which is taken from a 2012 paper
by Brenna M Henn , Lawrence Hon, J Michael Macpherson, Nick Eriksson, Serge Saxonov, Itsik Pe'er, and Joanna L Mountain. It multiplies the chances of
identifying cousins of different degree by the average number of cousins of
that degree:
However, the figures they use for the
number of cousins of each degree don't reflect the rate of growth of the
British population in the 19th century, which in England & Wales averaged
1.3% per annum, or about 50% per generation (for simplicity I've ignored
migration). You may recall that I mentioned
some time ago that Ancestry DNA had published some statistics on cousin numbers,
and I've incorporated them into my own table, which also draws on the table
from the 2012 paper. Note that I've added my own guesstimates for the numbers
of 7th and 8th cousins:
Degree of cousinship |
Expected amount of IBD (cM) |
Chance of detecting cousin |
Expected number of cousins |
Expected number of detectable cousins |
1 |
900 |
100 |
5 |
5 |
2 |
225 |
100 |
28 |
28 |
3 |
56 |
89.7 |
175 |
157 |
4 |
14 |
45.9 |
1570 |
720.6 |
5 |
3.5 |
14.9 |
17300 |
2577.7 |
6 |
0.88 |
4.1 |
174000 |
7134 |
7 |
0.22 |
1.1 |
1000000 |
11000 |
8 |
0.055 |
0.24 |
5000000 |
12000 |
9 |
0.014 |
0.06 |
||
10 |
0.0034 |
0.002 |
These are very different figures, and that
final column is very revealing - it suggests that we're typically going to find
around 30 to 40 times as many 5th to 8th cousins as we are 1st to 4th cousins,
even though the chances of our more distant cousins sharing detectable segments
of DNA with us is very much lower.
I should warn you that these numbers are
only rough estimates, but they demonstrate why, if you have British ancestry -
as most readers of this newsletter do - you're likely to get so many distant
matches.
There's another factor that can increase
the chances of finding distant cousins - intermarriage in small communities.
Communities can be isolated geographically, ethnically, by religion, or by
social class - the last two explaining why three of the major players in the
Great War were countries whose heads of state were grandsons of Queen Victoria
(indeed, according to this BBC article,
in 1914 there were 7 of her direct descendants and 2 of her Coburg relations on
European thrones).
At one level matches with distant
cousins are a bad thing - it makes it much more difficult to work out who the common ancestors were - but, at the same time it
ultimately allows us to use DNA to research back further than would otherwise
be possible.
Why place names are
more important than surnames
As we've already noted, the chances of
finding surnames we recognise in the trees of our DNA cousins are small - so
what other options do we have?
People didn't move around as much in
earlier centuries - so geography is likely to be a much better guide than
surnames (which, on average, changed every other generation). Even when people
did move, they tended not to move far, at least until the Industrial Revolution
took hold - so you're much more likely to find that your DNA cousins have the same
places in their tree than the same surnames. Usually it won't be the same
parish, but knowing that your cousin's ancestors and your own were (say) 20
miles apart in the 18th century strongly suggests where in your tree the common
ancestor is likely to be found.
Note:
you're fairly unlikely to find the same parish mentioned in both trees - for
the simple reason that when we run into a 'brick wall' it's usually because we
can't find our ancestor's baptism in the parish where they married (or their
marriage in the parish where their children were baptised).
Of course, the common ancestor could be
several generations back from your 'brick wall' ancestor, so even when you've
found a geographical overlap you've still got research to do - but at least
you've got a better idea where to look. And geography isn't the only tool in
your armoury - the techniques I describe in the following articles are even
more powerful!
How common matches
can lead to common ancestors
Let's suppose that you've tested with
Family Tree DNA (or transferred your results from another company) - if so,
you'll have a page that looks something like this:
The graphic shows my top 5 matches, one
of which is with my brother, and two with known cousins (the fourth match is
with a cousin who I found using DNA). If I click any one of my matches, so that
it is highlighted in yellow, then click the in
Common With tab, I'll get a list of all the DNA cousins that I share with
the highlighted cousin.
Sometimes when you do this, one of those
common matches will be with a known cousin - in which case you can deduce, with
a reasonable amount of certainty, that the connection
with your DNA cousin is on one of the lines you share with your known cousin.
Note:
when I refer to a known cousin in this context I'm talking about someone whose
connection to you is known - it could be someone you've known since childhood,
someone you found through LostCousins (or another site), or someone you've
found previously as a result of following up your DNA matches.
But it's not necessary to share your DNA
cousin with a known cousin in order to learn more about how you match - in most
cases you'll be able to make deductions by looking at the surnames and
geographical locations in the trees that have been uploaded by your common
matches. The extent to which analysing your common matches will help inevitably
varies - for example, if you both match someone who lives in Germany and has
only German ancestors, you can be reasonably confident that you and your DNA
cousin have shared German ancestry (although your most recent common ancestor
might have lived elsewhere).
Usually it won't be quite so obvious -
nevertheless, if you combine all the hints from the different matches you may
well have a pretty good idea where the match is, even if only geographically.
Note:
although I used Family Tree DNA in the example above, you can carry out similar
comparisons at GEDmatch.
Why you should ask your
cousins to test...
The more known cousins who have tested,
the more likely it is that when you find a new DNA match, you'll share that
match with one of your cousins.
But which cousins should you ask to take
an autosomal DNA test - should they be close cousins or distant cousins?
When distant cousins test you're
narrowing down the possibility of shared matches. For example, a 4th cousin -
someone who shares your 3G grandparents - shares only 1/16th of your tree, so
when you find a DNA cousin who matches you both you can reasonably assume that
the match is on that part of your tree. But the chance that you'll both have a
detectable match with a specific individual who is (say) a 6th cousin to both
of you is quite low - probably less than 1%. On the other hand, you'll have
quite a lot of matches with 6th cousins, and roughly 1/16th of those will be
with from the relevant part of your tree - so the odds aren't as bad as they first
look.
So when distant cousins test your chance
of finding common matches is lower - but the value of the matches is higher.
The closest relative I could ask to test
was my brother, and whilst you might think that because we share the same
parents we get all the same matches, that isn't the
case. In fact my brother only shares about half of my matches - so between us
we have around 50% more matches than either of us does individually. This is
because between us we have 50% more DNA - for whilst we have the same parents,
we each inherited a different subset of their DNA.
Knowing that my brother and I match with
the same DNA cousin tells me nothing about where on our tree the common
ancestor might be (because my brother and I have the same ancestors). But it
does increase the total number of matches by half.
... and why you should
ask your cousins if they have already tested!
Combining the statistics that various
DNA testing companies have released about the number of autosomal tests they
have carried out suggest that over 3.5 million people around the world have
tested. Although this is a very small percentage of the world population, I
would guess that most of them live in the English-speaking world, and that most
of them are researching their family tree.
If that's the case, then bearing in mind
that Ancestry had less than 2.5 million subscribers, there's a good chance that
many of the cousins you already know through your research have already tested.
Now, I don't know about you, but I don't
keep in touch with my 'found cousins' as regularly as I probably should. If I
have their postal address I send them a Christmas card and a 'round robin', but
we tend to email each other very intermittently.
The point I'm making is that you
probably have several cousins who have taken autosomal DNA tests, but haven't
thought to mention it to you. Why not get in touch with ALL of your cousins who
are researching their tree to find out whether they (or a close relative) have
tested?
Time to update your My Details page
There's a new field on your My Details page where you can specify
whether you've taken an autosomal DNA test or not - or if it's something you're
seriously considering. Right now the only people who will see what you enter
are you and me, but at some point in the future that information will also be
available to the other LostCousins members you've been matched with:
This will make it far easier to keep
track of which cousins have tested - and it's yet one more reason* to recommend
that cousins you find through other sites join LostCousins!
I recently pointed out the very
different ethnicity estimates that my brother and I got from Family Tree DNA
(you can read the original article here).
I ended with the rhetorical question "We have the same parents, so how can
our ancestry be so different - of course, the answer is that it can't!",
which some people interpreted as if I'd written "how can our ethnicity
estimates be so different" (which isn't the same question at all).
The answer to the latter question is
simple - we inherited different parts of our parents DNA. And that's one of the
problems with ethnicity estimates, you'll get different
results depending which sibling tests.
You'll also get different results depending
which company you test with. When my brother tested with Ancestry they said he
was 20% Irish, though not one of the ancestors I've found so far comes from
Ireland - we don't even have any Catholic ancestors (which would be a possible
pointer towards Irish ancestry).
So which of the three ethnicity
estimates should we believe - well, my answer is NONE OF THEM! At best they'll
confirm what you already know, or think you know, at worst they'll send you off
on a wild goose chase. Testing your DNA in order to discover your ethnicity is
like buying a lottery ticket in the expectation of winning the jackpot - the
odds are against you.
Note:
there will be rare occasions when an ethnicity estimate helps to confirm or
deny a family rumour about an ancestor who married a slave, an Eskimo, or a Red
Indian princess (and yes, I do know we don't call them that any
more, so please don't write in!). But these instances will be few and far
between - like lottery jackpots.
If you’re looking at another researcher's
family tree at Family Tree DNA it's important to be aware that in 'Family View'
you're limited to 4 generations. I always switch to 'Ancestry View' which can
have up to 15 generations, and switch on 'Detailed View' so that I can see at a
glance where everyone was born.
More DNA tips &
techniques to come
The tips and techniques that I've
described so far are sufficient to get you started, but there's so much more to
DNA for the real enthusiast.
In future articles I'll be writing about
chromosome browsers, triangulation, and the possibility that your DNA matches
somebody else's purely by chance - which could mean that your DNA cousin isn't
a cousin at all.
Mother wins right to use
dead daughter's eggs
It's a very sad story - and it would
have been a controversial outcome whichever way it was decided. The Court of
Appeal has decided that a mother can take her late daughter's frozen eggs to
New York so that she can give birth to her own grandchild. You can read more
about this difficult case here,
on the BBC news site.
A new way to search for war
deaths
I've long used the Commonwealth War Graves
Commission site to search for relatives who died in the two World Wars, but
it was only this week that I realised that it was possible to search by place
name (by entering it in the Additional
information box). If you're interested in local history as well as family
history this could be very useful, but I'm sure there will be many other
applications - house history could be one.
However, my wife spotted that there's an
even better way to search the CWGC database by address at the WalesOnline website - not only can you search by street
name (as well as by the name of the town or village), the results show the
names and addresses, so you don't have to click on individual entries.
How Newfoundland soldiers
forged lasting link with Scotland
Eight hundred members of the
Newfoundland Regiment went into battle at Beaumont-Hamel on 1st July 1916, the
first day of the Battle of the Somme - but the next morning at roll call only
82 answered. You can read more about this tragic story and how Scotland is
involved in this BBC Scotland article.
By the way, this week I had the pleasure
of hearing Professor Gary Sheffield talk about his magisterial book Douglas
Haig: from the Somme to Victory, and I believe I managed to persuade
him to write an article for this newsletter.....
Britain's oldest person dies, aged 113
On Friday Gladys
Hooper, Britain's oldest person passed away - she celebrated her 113th
birthday on 18th January. Last year she was the oldest person in the world ever
to have a hip operation - at the age of 112!
By a sad coincidence, the oldest person
in the USA, Goldie
Michelson of Massachusetts, also died the same day - and she was also 113,
though she was a few months older than Gladys, having been born in 1902.
Still, I'm sure that most us would settle for living to 113 - I'd love to make it to
102 so that I can see myself on the 1951 Census!
New CEO takes over at
Findmypast
Jay Verkler,
former CEO of FamilySearch, has been appointed as interim CEO of
Findmypast. He takes over from Annelies van den Belt who has stepped down
after three years.
I've never knowingly met Mr Verkler, but he has a lot of experience - as CEO of
FamilySearch for a decade from 2002 to 2012 he oversaw the period of transition
from paper and microfilm to digital and online records. The announcement
suggests that his term of office will be just 6 months, but remembering back to
what his predecessor achieved in her first 6 months we could still see a lot of
changes!
One thing is missing from the
announcement - it doesn't mention that he's interested in family history. But
it's hard to imagine that with a background like his, he hasn't researched his own
family tree!
Try out Findmypast with
a half-price World subscription EXTENDED???
You can save 50% on a 1 month Findmypast
subscription when you follow the appropriate link below:
A World subscription can be used at any
of Findmypast's four sites, and gives access to all of Findmypast's 8 billion
records and newspaper articles with the
exception of the 1939 Register (which is only available to subscribers with
a 12 month Britain or World subscription).
If you're an Ancestry subscriber it's a
great opportunity to get access to a massive collection, which includes many
records which are exclusive to Findmypast - so you can fill in those annoying
gaps in your tree! But to make the most of your initial month you'll need to
hit the ground running, so see this article
from March which explains how to get the most out of the Findmypast site.
Of course there are some records, such as
censuses, that are also available at other sites but most of the record sets aren't
online at any other sites: for example, highlights from the British collection
include parish registers for Cheshire, Devon, Hertfordshire, Plymouth & West
Devon, Shropshire, Staffordshire, Westminster, most of East Kent, large parts
of Yorkshire, and much of Wales (all
exclusive to Findmypast). Findmypast also has partly indexed register images
for Lincolnshire, whilst Leicestershire and Rutland parish registers will
also be going online at Findmypast later this year.
You'll see this offer elsewhere - it isn't
exclusive to LostCousins - but you'll only be supporting LostCousins if you use
the links at the start of the article (if they don't work for you, see the
notes at the start of this newsletter).
Note:
your monthly subscription will be automatically renewed at the normal price
unless you change the auto-renew setting under Personal details in the My
Account section of the site. You can do this at any time.
How has Brexit affected
genealogists?
Even if you have no interest in politics
or economics, I'm sure you will have noticed that two weeks ago the UK voted to
leave the European Union. The most immediate effect was a big fall in the
British pound against other currencies - against the US dollar it set a 31-year
low.
For anyone outside the UK who is buying
subscriptions, certificates, books etc that are
priced in pounds this is really good news: for example, a LostCousins
subscription that would have cost US$15 last month will now cost about $13.
Similarly a birth, marriage or death certificate that would have cost you $14
last month will now cost about $12. There are also useful savings for
researchers who live in Canada, Australia, New Zealand, Ireland, mainland
Europe - indeed, just about anywhere other than the UK.
Unfortunately for those of us in the UK
(which means two-thirds of the readers of this newsletter) it works in reverse
- foreign holidays will cost more, and so will the cost of imports such as wine
from the New World, cheese from France and, inevitably, fuel (since oil is
priced in US dollars). And, if Genealogy
in the Sunshine goes ahead in Portugal next March, as I hope, that will
also be more expensive.
But since this is a DNA Special edition,
perhaps most relevant is the increase in the cost of Family
Finder tests, which will now cost around £91 including shipping, £10 more
than before. This is much closer to the cost of an Ancestry
DNA test, which costs £99 in the UK including shipping (though if you order
more than one test at the same time there is a £10 saving in the shipping
charge).
However, price isn't the only
consideration - there are other factors to take into account, such as the
limited functionality if you discontinue your Ancestry subscription (Family
Tree DNA doesn't have subscriptions), and the lack of tools to analyse your
matches.
Last month I had a good moan about BT,
who tried to shock me out of switching to Sky by telling me I could lose
"up to £194.28" in prepaid line rental (as you may recall, this was
totally untrue).
They eventually seemed to accept that I
was leaving, but then sent me a parting gift - an invoice for £67.92 for the
following month, which they promptly took from my bank account! So which part
of "I'm leaving" don't they understand?
But don't let BT's incompetence put you
off buying their BT8500 and BT8600 call-blocking phones - only today I
received an email from Judey in which she wrote:
"I'm thrilled with my BT call blocker phone which I got on your
recommendation."
Incidentally, there is little difference
between the two models - my advice is to buy whichever is cheapest in the
configuration you need (all include a digital answering machine, so you can
cancel 1571, saving over £2 a month). And if you're shopping on Amazon (as I
usually do), look at the deals offered by third-party sellers, especially for
items in damaged packaging.
If you like saving money and spending
money then you might want to consider signing up for a free one month trial of
Amazon Prime now, so that it includes Prime Day (12th July), when there are
lots of discounted deals. I bought several things from the UK site last year
(and yes, for the benefit of my wife - who is reading this - they were things I
actually needed!), so I'm hoping that this year will be just as good. You can
support LostCousins by using one of the following links to sign up for your
free trial:
Working from home I don't pick up many
colds, but when they do strike - as one did this week - they're pretty
devastating. Hopefully today will be the low point (but then I thought that
yesterday, and the day before); so if I'm a little slow responding to emails,
and perhaps a little grumpy, please forgive me - I feel awful!
The Findmypast offer above should have ended
on Friday 15th July, but when I checked today (12.49pm 16th July) it was still available
- so get in quick!
That's all for now - but if you're still
mystified by DNA, please see the series of introductory articles I published
last year:
Understanding DNA #1: your genetic
inheritance
Understanding DNA #2: mtDNA myths
Understanding DNA #3: the truth
behind DNA tests
Understanding DNA #4: how DNA is inherited
Understanding DNA #5: choosing the right test
Understanding DNA #6: choosing the
right company
Understanding DNA #6: choosing the
right company (continued)
Peter Calver
Founder, LostCousins
© Copyright 2016 Peter Calver
Please
do not copy any part of this newsletter without permission. However, you MAY
link to this newsletter or email a link to your friends and relatives without
asking for permission in advance - though why not invite them to join
LostCousins instead as standard membership, which includes this newsletter, is
FREE?