Decoding Kin: Genealogical Bioinformatics Solves Family Mysteries

WEBVTT
00:00:06.008 --> 00:00:11.304
Wagwan, everybody. Welcome to the Dis Afimi History podcast,
00:00:11.304 --> 00:00:17.114
where we'll be speaking about history and as well family history and how history
00:00:17.114 --> 00:00:19.064
relates in terms of Caribbean people,
00:00:19.994 --> 00:00:25.814
for the present as well as in the past and how in the past, what that does and
00:00:25.814 --> 00:00:30.164
brings forth for what we are going through at present and what we can learn
00:00:30.164 --> 00:00:33.684
from our history, from our family and take that moving forward.
00:00:33.684 --> 00:00:38.336
So I do hope you enjoy the podcast, and if you like it, please ensure to subscribe,
00:00:38.534 --> 00:00:40.350
like, and review. Thank you.
00:00:43.584 --> 00:00:48.574
I just want to thank you so much for coming on to the podcast and to talk about
00:00:49.014 --> 00:00:50.856
genetic bioinformatics.
00:00:51.129 --> 00:00:55.464
So for listeners who may be new to this field, can you just introduce yourself
00:00:55.464 --> 00:00:58.774
and explain the work you do through discovering new ancestors?
00:00:59.245 --> 00:01:03.413
Certainly, and thank you for having me. Great. I'm a genetic genealogist,
00:01:03.751 --> 00:01:08.104
although I've also certainly done a lot of conventional genealogy research in
00:01:08.104 --> 00:01:09.618
researching my own family.
00:01:09.966 --> 00:01:14.061
Most of the client work I would do for people involves genetics.
00:01:14.537 --> 00:01:18.574
In addition to doing research with the conventional tools that,
00:01:19.054 --> 00:01:25.164
many people use, I've also spent the past five, six years developing my own
00:01:25.164 --> 00:01:31.213
tools and methods in this field, which I refer to as genealogical bioinformatics.
00:01:31.689 --> 00:01:35.437
And I do this through my business, Discovering New Ancestors.
00:01:35.756 --> 00:01:41.924
To date, I have filed five patents for processes relevant to bioinformatics,
00:01:41.924 --> 00:01:44.005
and I have a sixth process in development.
00:01:44.545 --> 00:01:51.014
And I found these tools are very useful in addressing longstanding problems
00:01:51.014 --> 00:01:55.812
that genetic genealogy has been unable to talk about through conventional methods,
00:01:56.288 --> 00:01:59.966
as well as assisting me in my own personal research.
00:02:00.256 --> 00:02:03.125
Many of these processes were motivated by.
00:02:03.819 --> 00:02:08.806
Problems I had in my own research, which I needed to address for my own work.
00:02:09.381 --> 00:02:14.576
No, thank you so much. And again, what is exactly genetic bioinformatics?
00:02:14.576 --> 00:02:17.646
And, you know, how would you be able to explain it to someone that,
00:02:17.646 --> 00:02:20.707
you know, for everyday genealogists that they can be able to understand?
00:02:21.113 --> 00:02:26.156
Certainly, certainly. Genealogical bioinformatics basically takes everything
00:02:26.156 --> 00:02:28.791
we know about genetic genealogy.
00:02:29.458 --> 00:02:34.452
And to that, it adds elements of what we call information engineering,
00:02:34.893 --> 00:02:38.597
which are basically different ways of looking at a familiar problem,
00:02:39.021 --> 00:02:44.866
looking at these problems in ways that a computer might be able to more formally
00:02:44.866 --> 00:02:48.663
understand them in terms of using elements of mathematics
00:02:49.034 --> 00:02:54.636
and logic to take information and things that we almost intuitively understand.
00:02:55.396 --> 00:03:01.466
As a genealogist, but looking at them in a way that a computer or these days
00:03:01.466 --> 00:03:05.036
an AI system might be better able to understand also.
00:03:05.564 --> 00:03:09.216
And so we take everything we know about genetic genealogy, and none of that's
00:03:09.216 --> 00:03:14.566
wrong, but we take those methods and we marry them up with ways of looking at
00:03:14.566 --> 00:03:18.196
information, certain things that we would call pattern recognition,
00:03:19.536 --> 00:03:24.122
something called data mining, basically ways of looking at huge amounts of data,
00:03:24.633 --> 00:03:26.078
and trying to make sense of them.
00:03:27.570 --> 00:03:31.016
And, you know, what first drew you to this area of research and what made you
00:03:31.016 --> 00:03:34.419
decide to become such a strong advocate and public voice for this?
00:03:35.185 --> 00:03:40.369
Well, I found that when researching my own family tree and ancestry,
00:03:40.764 --> 00:03:45.332
there were questions I had and brick walls, which weren't,
00:03:45.731 --> 00:03:49.553
which were basically, again, I guess, brick walls, things that couldn't be solved
00:03:49.820 --> 00:03:50.905
immediately and easily.
00:03:51.422 --> 00:03:56.002
And yet I knew that there had to be an answer to some of these questions.
00:03:56.402 --> 00:04:01.975
In some cases, I had ancestors for whom I didn't have any really good documentation
00:04:02.411 --> 00:04:06.827
and they didn't appear in other people's family trees and there weren't really
00:04:06.827 --> 00:04:08.860
any good conclusive records about them,
00:04:09.336 --> 00:04:11.861
in terms of who their parents or cousins were.
00:04:12.343 --> 00:04:16.680
But yet I knew that, okay, if I've tested and my cousins have tested,
00:04:17.069 --> 00:04:21.527
I must have matches which connect me through this person.
00:04:21.852 --> 00:04:24.098
I just don't know who they are or where they are.
00:04:24.644 --> 00:04:30.177
And by thinking about that, I began to try to work out some methods that would
00:04:30.177 --> 00:04:36.847
allow me to find those people and isolate those people and at least be able to sort of look.
00:04:38.127 --> 00:04:42.427
At my whole set of DNA matches and say, well, these ones I know and these ones
00:04:42.427 --> 00:04:48.512
I don't know, and these ones here, these have to be from my mystery ancestor. And,
00:04:49.167 --> 00:04:54.427
What I found was that genetic genealogy was very good at working with close
00:04:54.427 --> 00:05:01.369
matches and even DNA matches that are of a medium and to medium close range.
00:05:01.665 --> 00:05:04.457
But then when you began to work with really distant matches,
00:05:04.742 --> 00:05:10.177
I found that I needed to find ways to work with a lot of people's data,
00:05:10.177 --> 00:05:14.402
my data, my cousin's data, and really try to sort of make sense of it
00:05:14.808 --> 00:05:17.607
in a way that went beyond what one person could do.
00:05:17.607 --> 00:05:22.761
I needed to sort of bring a computer in to start working with all of these points of data.
00:05:23.386 --> 00:05:26.967
And that's when I really began to sort of say, OK, well, if I'm going to do
00:05:26.967 --> 00:05:30.642
this, I need to sit down and really think about what I'm doing.
00:05:31.008 --> 00:05:36.387
And I need to I'm going to need to build some tools of my own which can take
00:05:36.387 --> 00:05:40.586
these bits of information and make sense out of them. And that's really where
00:05:40.923 --> 00:05:42.796
the tools began to evolve.
00:05:43.150 --> 00:05:47.377
The first tool I filed a patent for was in the fall of 2021.
00:05:48.805 --> 00:05:54.067
And then things grew from there. And how does, I guess, genetic bioinformatics
00:05:54.067 --> 00:05:58.124
differ from traditional genealogical methods?
00:05:58.542 --> 00:06:03.495
Good question. Basically, traditionally with genetic genealogy,
00:06:03.814 --> 00:06:07.628
we look at how much DNA you would share with someone.
00:06:08.367 --> 00:06:12.667
And there are fairly well-established guidelines for what you would share with
00:06:12.667 --> 00:06:18.263
a child or a parent or an aunt or an uncle or a cousin or a second cousin.
00:06:18.891 --> 00:06:25.015
And we take that information, and then we also work with things that we understand
00:06:25.015 --> 00:06:30.037
as genealogists, but we don't necessarily lend themselves to hard numbers.
00:06:30.385 --> 00:06:35.087
For instance, if I talk to you about your maternal cousin, you know intuitively
00:06:35.302 --> 00:06:38.112
who they would match. They would match people on your mother's side,
00:06:38.234 --> 00:06:41.943
and who they wouldn't match. They wouldn't match people on your father's side so much.
00:06:42.755 --> 00:06:47.931
We look at things like that, and we express those things in mathematical terms,
00:06:48.181 --> 00:06:51.045
terms which a computer system can better understand.
00:06:51.347 --> 00:06:55.695
And the reason we do that is because we're not dealing just with our close matches,
00:06:56.055 --> 00:07:01.055
but if a person takes a DNA test, they often have thousands of matches.
00:07:01.055 --> 00:07:03.944
They'll have 20,000, they'll have 40,000 matches.
00:07:04.263 --> 00:07:09.665
If they come from a community, which might be an island population or something
00:07:09.665 --> 00:07:13.815
that has a group of people that has a dogomy, then you might have 60,
00:07:13.815 --> 00:07:15.927
80, 100,000 DNA matches.
00:07:16.298 --> 00:07:20.971
And that's really much too much information for a person to deal with on their own.
00:07:21.314 --> 00:07:26.422
And so by treating not just looking at a single point of information,
00:07:26.712 --> 00:07:30.346
but rather looking at the relationships between matches and points,
00:07:30.700 --> 00:07:35.775
we begin to go beyond just what genetic genealogy can do and look at,
00:07:36.510 --> 00:07:41.328
the whole, the totality of your DNA matches as basically a unit of information.
00:07:41.810 --> 00:07:47.515
And that if we take that unit of information and we compare it to other people's
00:07:47.515 --> 00:07:50.274
results, we can then devise tests
00:07:50.506 --> 00:07:56.655
that can tell us more things about our matches based on how our matches.
00:07:57.533 --> 00:07:58.886
Respond to those tests.
00:07:59.211 --> 00:08:04.522
And so we begin to, we can sort of read information and connect the dots that way.
00:08:05.045 --> 00:08:08.595
Okay. And then for, I guess, for people researching their own family history,
00:08:08.595 --> 00:08:15.323
what kinds of questions can this approach help answer that conventional records sometimes cannot?
00:08:16.089 --> 00:08:22.006
Well, one thing to consider is that when we look at, when we test our DNA.
00:08:23.375 --> 00:08:28.557
It's a given that we'll have DNA matches connected through both of our mother and father.
00:08:28.975 --> 00:08:33.625
And we'll have DNA matches slightly more distant. That'll be cousins that connect
00:08:33.625 --> 00:08:37.067
to one or any of our four grandparents.
00:08:37.625 --> 00:08:42.833
And then second cousins will be connected to us through any of our great grandparents.
00:08:43.204 --> 00:08:48.265
But when you move beyond that, what we find is that we don't necessarily have
00:08:48.265 --> 00:08:52.465
matches that connect us with each of our great-great-grandparents.
00:08:52.465 --> 00:08:57.675
Some of that DNA may have washed out, or we may have inherited some DNA from
00:08:57.675 --> 00:09:03.139
a second-great-grandparent, and our third cousins may have inherited other parts of their DNA.
00:09:03.400 --> 00:09:08.825
And so even though we're biologically related, we don't match in DNA to quite
00:09:08.825 --> 00:09:10.115
the way we would expect to.
00:09:10.533 --> 00:09:18.265
And what bioinformatics can do is basically allow us to map out where our DNA
00:09:18.265 --> 00:09:22.339
matches lie, where we have received DNA and where we haven't.
00:09:22.623 --> 00:09:26.605
So that rather than choosing a certain ancestral line and saying,
00:09:26.605 --> 00:09:28.066
oh, this is what I want to find.
00:09:28.339 --> 00:09:34.411
Instead, we can ask the question, what DNA do I have and where does that DNA lead me?
00:09:34.736 --> 00:09:37.464
It might not be where we often expect.
00:09:37.917 --> 00:09:41.174
It may not be the line we're most interested in.
00:09:41.522 --> 00:09:45.975
Often what people will do is have a family name, and often it's a father's name
00:09:45.975 --> 00:09:48.912
and the father's father, and people want to research that.
00:09:49.289 --> 00:09:53.016
And sometimes that's great because they have inherited that DNA,
00:09:53.312 --> 00:09:57.468
but other times they may have inherited DNA from the father's mother's family.
00:09:57.904 --> 00:10:03.235
And these bits of DNA, again, it's not even obvious where they come from until
00:10:03.235 --> 00:10:07.925
we begin to look at the big picture and correlate what we have with what our
00:10:07.925 --> 00:10:10.583
cousins have and begin to see that, oh.
00:10:11.257 --> 00:10:15.052
Whatever questions we want to answer, this is where we have to look.
00:10:15.052 --> 00:10:18.504
And it's not necessarily where we thought we were going to look.
00:10:19.124 --> 00:10:24.152
Yeah, exactly. And, you know, for some genealogists, do they have to have a
00:10:24.152 --> 00:10:29.852
scientific background to kind of incorporate bioinformatics into their own research?
00:10:30.601 --> 00:10:35.750
Well, the promising thing about these tools is that.
00:10:37.072 --> 00:10:43.662
Basically the ideas at their heart are scientific ideas, but the tools I've
00:10:43.662 --> 00:10:51.005
built will allow you to do the work and have the tool do a lot of the processing and lifting.
00:10:51.482 --> 00:10:56.602
What you would still have to do at this point is you would have to harvest or
00:10:56.602 --> 00:10:58.566
make a downloadable copy of
00:10:58.746 --> 00:11:02.839
the full set of your matches and the full sets of your cousin's matches.
00:11:03.094 --> 00:11:07.162
And that's a lot of work, but it's still something that's doable on with the
00:11:07.172 --> 00:11:09.838
computer if you are inclined to do that.
00:11:10.472 --> 00:11:15.620
Ultimately, what people, really the way these tools will benefit the most people
00:11:15.992 --> 00:11:20.549
is if the methods are actually eventually built into the back-end reporting
00:11:20.717 --> 00:11:26.472
of a DNA testing provider so that really you can ask the questions and the database
00:11:26.832 --> 00:11:31.652
will give you the answers, but you don't need to do any of the science any more
00:11:31.652 --> 00:11:35.852
than, let's say right now, when you take a test, you log in and you read your
00:11:35.852 --> 00:11:39.400
matches, but you don't have to do any science to do that. You simply,
00:11:40.277 --> 00:11:44.108
read what your answers are in terms of who do I match, who do I not match,
00:11:44.590 --> 00:11:48.422
and all the hard work is done in the cloud.
00:11:48.787 --> 00:11:55.496
And that's ultimately where these tools will do their best work for the most greatest people.
00:11:56.953 --> 00:12:02.243
The thing I'm doing right now often is telling people that these tools exist
00:12:02.643 --> 00:12:04.193
so that there's an awareness.
00:12:04.659 --> 00:12:08.937
When people often are aware that they have a problem, they don't necessarily
00:12:08.937 --> 00:12:10.117
know what can be done about it.
00:12:10.117 --> 00:12:15.517
And so to try and sort of educate people and inform them that this is what's
00:12:15.517 --> 00:12:20.952
next, this is what can be done, you need to just sort of make them aware that it's even possible.
00:12:21.585 --> 00:12:26.117
No, that's good to know. And in terms of with, I guess, with the role of DNA,
00:12:26.117 --> 00:12:29.497
how does that play within genealogical bioinformatics?
00:12:29.497 --> 00:12:34.757
And how do you kind of balance the genetic evidence with archival and,
00:12:34.757 --> 00:12:37.884
of course, documentary records that you would obtain?
00:12:38.940 --> 00:12:45.007
Well, the DNA evidence is basically our sort of what I would describe as our source materials.
00:12:45.007 --> 00:12:52.337
We start with that, and then your DNA matches, your raw DNA data are the raw
00:12:52.337 --> 00:12:54.120
materials that go into the system.
00:12:54.462 --> 00:12:59.137
There is, of course, no such thing as genetic proof. DNA won't prove anything.
00:12:59.137 --> 00:13:03.011
It's only a form of evidence similar to archival records.
00:13:03.388 --> 00:13:07.701
And so all of these things have to be balanced out to make sense with each other.
00:13:08.026 --> 00:13:11.179
The DNA evidence, at least, we're able to read and interpret.
00:13:11.776 --> 00:13:16.235
And with either conventional genetic genealogy or with bioinformatics,
00:13:16.542 --> 00:13:21.709
we can interpret that information and get suggestions and answers.
00:13:22.040 --> 00:13:27.264
But then still, we need to do the research with that evidence to make sense of things.
00:13:27.578 --> 00:13:33.813
So this is really sort of allowing us to get more evidence and to often as not,
00:13:34.463 --> 00:13:39.200
really see what's, what information is already there in our DNA matches,
00:13:39.537 --> 00:13:41.923
and then to try to reconcile that with.
00:13:42.896 --> 00:13:46.699
With what makes sense, what is actually reality. Yeah.
00:13:47.047 --> 00:13:50.897
And I think you mentioned it in a previous answer where, you know,
00:13:50.897 --> 00:13:54.857
there are certain communities or populations such as other diasporic or enslaved
00:13:54.857 --> 00:14:00.002
dissented communities where this work can be especially meaningful for them.
00:14:00.647 --> 00:14:06.387
Yes, absolutely. Some of basically, as I mentioned, one of the things that I
00:14:06.387 --> 00:14:11.281
find very useful about bioinformatics is that it can answer questions or address
00:14:11.468 --> 00:14:14.323
longstanding problems that genetic genealogy has.
00:14:14.683 --> 00:14:19.562
And one of those problems that some of my processes are better at dealing with
00:14:19.869 --> 00:14:21.252
is something we call endogamy,
00:14:21.658 --> 00:14:27.367
which basically, again, often happens in historically isolated or marginalized
00:14:27.367 --> 00:14:32.107
communities where people have either been forced because of geography, let's say, to
00:14:32.487 --> 00:14:36.767
marry within a certain group, again, if you were all on a certain island, or again, socially,
00:14:37.167 --> 00:14:41.537
often enslaved people or people at the margins of society were obliged,
00:14:41.537 --> 00:14:44.187
you know, didn't have much control over who they.
00:14:45.047 --> 00:14:46.771
Could be involved with.
00:14:47.338 --> 00:14:51.531
And so, again, those are situations where,
00:14:51.990 --> 00:14:56.517
people, even if they take a DNA test, will find that the results don't often
00:14:56.517 --> 00:15:03.118
make sense because they share more DNA with a closed group of people than we would expect them to.
00:15:03.553 --> 00:15:08.957
And often is not historically what we've tried to do within genetic genealogy
00:15:08.957 --> 00:15:14.387
on its own, was to either work with the results as best we can,
00:15:14.887 --> 00:15:21.640
or to look for results where we have a match, which has a large number of segments to it.
00:15:22.633 --> 00:15:28.583
Or basically disregard small matches, especially in endogamy situations,
00:15:28.931 --> 00:15:31.160
because those matches tend to be less reliable.
00:15:31.683 --> 00:15:36.390
And what bioinformatics can do is it can actually,
00:15:37.048 --> 00:15:43.710
create an individualized profile for each of us based on who we match and who
00:15:43.710 --> 00:15:49.350
we don't match and where our genetic pileups may exist so that, in fact, even,
00:15:50.010 --> 00:15:56.186
DNA matches from endogamous situations can be made to more closely resemble
00:15:56.482 --> 00:15:58.659
what we expect to receive in a DNA test.
00:15:58.944 --> 00:16:05.480
And so, on a very individual, personalized basis, our DNA results can make sense,
00:16:05.480 --> 00:16:08.493
or at least we can understand them better through these tools.
00:16:09.610 --> 00:16:13.620
Absolutely. And, you know, what would be some of the common misconceptions that
00:16:13.620 --> 00:16:17.490
people have about, of course, DNA and then genealogy? research?
00:16:18.218 --> 00:16:24.859
Well, a common misconception, which I mentioned briefly, was that DNA can prove things.
00:16:25.161 --> 00:16:31.906
In fact, DNA is a type of evidence, and what we do with that evidence really constitutes our proof.
00:16:32.570 --> 00:16:36.190
In some cases, the proof may be obvious in terms of, you know,
00:16:36.190 --> 00:16:37.992
is this person my parent or not?
00:16:38.460 --> 00:16:40.240
In other cases, however, especially
00:16:40.240 --> 00:16:45.204
as we get more distant with our DNA matches, we really do have to,
00:16:45.856 --> 00:16:50.743
look at what's logical and what's possible and eliminate what isn't possible
00:16:51.080 --> 00:16:53.690
in order to find something resembling the truth.
00:16:53.690 --> 00:16:56.669
But there's still a lot of research and a lot of interpretation there.
00:16:57.381 --> 00:17:06.080
Often people have the misconception that genetic testing can tell us about matches
00:17:06.080 --> 00:17:12.027
that occurred hundreds and hundreds of years ago, and that's not always possible. But.
00:17:12.992 --> 00:17:18.931
Again, the more we work with these advanced tools, the better we can interpret
00:17:18.931 --> 00:17:20.771
what information we do have.
00:17:21.656 --> 00:17:26.428
And so, you know, DNA doesn't have all the answers. And I think that's often a misconception.
00:17:26.794 --> 00:17:32.651
But the more people we test and the more we look at our own results carefully,
00:17:32.976 --> 00:17:34.731
the more information we can find.
00:17:35.031 --> 00:17:39.401
So there are always possibilities, I like to say. Yeah, absolutely.
00:17:39.751 --> 00:17:43.601
Absolutely. And would you be able to share, I guess, a personal story or a discovery
00:17:43.601 --> 00:17:49.173
from your own journey with genealogical bioinformatics that deeply impacted you?
00:17:49.730 --> 00:17:57.327
Certainly, certainly. I mean, one of the tools, the first tool I built, which was in 2021,
00:17:57.988 --> 00:18:05.607
It is a process which I call CMA, which in the long term stands for Correlated Multiphasic Analysis.
00:18:06.008 --> 00:18:11.513
And what it basically does is it allows you to take your DNA matches and the
00:18:11.513 --> 00:18:14.419
DNA matches of your known family members.
00:18:15.487 --> 00:18:20.493
And it allows you to work with these matches and the relationships of you and
00:18:20.493 --> 00:18:26.543
your cousins to find more distant, basically to make sense of more distant matches
00:18:26.543 --> 00:18:27.742
than you might otherwise do.
00:18:28.177 --> 00:18:34.353
One of the problems with a small DNA match is that it may not be,
00:18:36.173 --> 00:18:40.973
a match which is generally, well, excuse me, a match which is generally connected
00:18:40.973 --> 00:18:45.653
to us through a distant ancestor, but it may be something that just occurs to
00:18:45.653 --> 00:18:48.963
us because our ancestors came from a particular place.
00:18:49.833 --> 00:18:54.617
And so everyone who comes from a certain island has this segment of DNA,
00:18:54.936 --> 00:19:00.788
or everyone who comes from Scotland or India or somewhere else has this particular type of match.
00:19:01.275 --> 00:19:06.773
And by correlating our matches with our cousins and other known family members,
00:19:07.092 --> 00:19:12.043
we're able to eliminate what we would call false matches and look at matches
00:19:12.043 --> 00:19:13.306
that have greater significance.
00:19:13.706 --> 00:19:18.733
So in my particular case, something I was interested in researching is my great,
00:19:18.733 --> 00:19:22.010
great, great grandmother. I didn't know much about her family.
00:19:22.306 --> 00:19:26.890
She was British and her family name was Jones, which was terribly common.
00:19:27.801 --> 00:19:32.690
And I wanted to learn more. I knew that there had to be, intuitively,
00:19:32.690 --> 00:19:38.378
I felt there had to be some answers to be found that didn't exist in the documents I was seeing.
00:19:39.116 --> 00:19:45.760
And because the surname was so common, it would be easy to find answers that
00:19:45.760 --> 00:19:49.901
without really being able to test or prove as to whether they were really correct.
00:19:50.759 --> 00:19:54.689
But I wanted to find more information about this ancestor's family.
00:19:55.148 --> 00:20:01.660
And what I did is I used my tool to basically look at my DNA matches and the
00:20:01.660 --> 00:20:08.130
DNA matches of about 20 other cousins and found that basically if I put all
00:20:08.130 --> 00:20:09.916
of those data points together,
00:20:10.537 --> 00:20:19.424
I could look and see what DNA matches were shared most by my cousins and myself.
00:20:19.836 --> 00:20:25.088
And some of those matches were people known to me and, and those things made, made perfect sense.
00:20:25.559 --> 00:20:29.771
But if I went down the list of who matched me and my cousins the most,
00:20:30.258 --> 00:20:36.733
I eventually came to one person who matched about 14 of my 20 cousins.
00:20:37.594 --> 00:20:42.522
But yet was a very small match and not someone whose connection was in any way obvious to me.
00:20:43.854 --> 00:20:48.854
I took that person and, you know, sort of messaged them and looked to see who
00:20:48.854 --> 00:20:55.054
they might be more closely related to and found the first cousin and a son of
00:20:55.054 --> 00:20:56.418
theirs who had also tested.
00:20:56.761 --> 00:21:00.994
But nothing about that was really very helpful either because I could see their
00:21:00.994 --> 00:21:04.458
immediate family, but I really couldn't see how they connected to my family.
00:21:05.374 --> 00:21:09.037
What I ended up doing was building another tool,
00:21:09.639 --> 00:21:14.674
that allowed me to take the full set of my mystery person's matches,
00:21:15.274 --> 00:21:21.444
and compare it back to my cousin's matches and my own and generate a small list
00:21:21.444 --> 00:21:27.279
of people who matched my family, but matched the mystery person more closely.
00:21:27.808 --> 00:21:33.124
And that list of six or eight people, I was then able to work up and build up
00:21:33.124 --> 00:21:36.805
their family trees through conventional genealogy research.
00:21:37.316 --> 00:21:41.794
And what I found is that they were all descended from a man named Christopher
00:21:41.794 --> 00:21:45.699
Bowman and his wife, whose family name was also Jones.
00:21:46.163 --> 00:21:52.184
And so here now was a family who basically I ended up finding 30 of them all
00:21:52.184 --> 00:21:57.544
together who had tested, who were all descended from the wife of Mr.
00:21:57.544 --> 00:22:00.840
Bowman, whose wife's name was Jones.
00:22:01.188 --> 00:22:06.523
And indeed, my great-great-great-grandmother, whose name was Susanna Jones,
00:22:06.750 --> 00:22:08.176
married a man who was a baker.
00:22:08.553 --> 00:22:13.122
And sure enough, this other woman, whose name was Christiana Jones.
00:22:14.034 --> 00:22:15.698
Her husband was also a baker.
00:22:16.133 --> 00:22:20.844
And in fact, she was born in London, England around the same time I would expect
00:22:20.844 --> 00:22:24.540
a a younger sister of my great-great-great-grandmother to have been born.
00:22:24.896 --> 00:22:30.544
And so now, even though I had this very common Jones family, I also now had,
00:22:31.296 --> 00:22:36.207
my great-great-great-grandmother, and a woman who I very strongly believe would be her sister.
00:22:37.115 --> 00:22:42.495
And this was all sort of there in the DNA matches, but I was able to find this
00:22:42.495 --> 00:22:49.615
just by using these tools and these techniques, which could at least lead me in the right direction.
00:22:51.155 --> 00:22:54.169
Ultimately, you still have to do a lot of conventional work and interpretation.
00:22:54.772 --> 00:23:00.962
But whereas myself and each of my cousins might have had 40,000 DNA matches,
00:23:01.327 --> 00:23:04.404
And so you're looking at 800,000 points of data.
00:23:04.793 --> 00:23:08.745
These tools were at least able to tell me, look at the family trees of these
00:23:08.745 --> 00:23:15.431
eight people who have tested out of 800,000 and build up their trees to see what line they share.
00:23:15.704 --> 00:23:18.136
And this is the ancestor you want to look at.
00:23:18.624 --> 00:23:24.695
The same tool also led me to another family group, which appears to share matches
00:23:24.695 --> 00:23:28.545
with both of the other groups I've talked about, but is probably descended from
00:23:28.545 --> 00:23:31.173
a third person from the Jones family.
00:23:31.627 --> 00:23:36.085
And so what this, of course, allows me to do is that rather than just looking
00:23:36.085 --> 00:23:40.007
for a Susanna Jones, who could be any number of people in England,
00:23:40.419 --> 00:23:44.053
I'm now looking for a Susanna Jones, who has a sister, Christiana,
00:23:44.482 --> 00:23:46.659
and very likely another descendant.
00:23:47.031 --> 00:23:51.575
And this allows us at least to take the DNA evidence and say,
00:23:51.575 --> 00:23:56.475
well, if anything exists, it needs to be documented with these names.
00:23:56.621 --> 00:23:59.680
And now we can bring in conventional research.
00:24:00.946 --> 00:24:04.826
And work with that. So that was something that was very exciting to be able
00:24:04.826 --> 00:24:11.889
to see that these very distant matches, again, we're looking at people who were born in the late 1700s,
00:24:12.561 --> 00:24:17.266
that their descendants have tested, but that because the matches are so small
00:24:17.266 --> 00:24:20.266
and so distant, you really need good evidence.
00:24:20.266 --> 00:24:26.830
You need to be able to say that this person matches, again, 12 of my 20 cousins
00:24:27.091 --> 00:24:32.095
in order to really say this is a significant small match and not just an accident.
00:24:32.432 --> 00:24:36.553
And then by working it from there, we begin to make sense of things.
00:24:37.116 --> 00:24:39.986
Well, that is just so fascinating because I know if it was just looked up by
00:24:39.986 --> 00:24:42.370
traditional, that would take probably years to even...
00:24:43.200 --> 00:24:46.553
Kind of weave that all together. But the fact that you were able to use those
00:24:46.553 --> 00:24:49.863
tools is just amazing. And then the other question is, I guess,
00:24:50.456 --> 00:24:55.783
what ethical considerations should researchers keep in mind when working with
00:24:55.783 --> 00:24:58.188
genetic information and family histories?
00:24:58.792 --> 00:25:03.083
Certainly. You know, a lot of the same concerns that would normally apply with,
00:25:03.543 --> 00:25:09.913
regular genetic genealogy apply here, that you want to be able to basically
00:25:09.913 --> 00:25:13.666
ethically have the consent of the people whose DNA matches you're working with.
00:25:14.183 --> 00:25:19.063
When I'm working with my own family members' matches, they understand that I'm
00:25:19.063 --> 00:25:23.743
going to information and download it to my computer and,
00:25:24.363 --> 00:25:29.300
work with it in my tools to find new connections and new bits of information.
00:25:29.736 --> 00:25:34.828
If I'm reaching out to people, as I did with my mystery match descended from,
00:25:35.333 --> 00:25:39.313
my great-great-great-grandmother's sister, then I explain to them that this
00:25:39.313 --> 00:25:41.404
is why, this is what I'm asking for.
00:25:41.660 --> 00:25:44.231
I'd like you to share your DNA matches with me.
00:25:44.609 --> 00:25:48.033
I'm going to download them and put them into my tool. And again,
00:25:48.033 --> 00:25:53.123
it all comes down to informed consent that people are going to give us their
00:25:53.123 --> 00:25:56.241
information, share their personal data with us.
00:25:56.589 --> 00:26:01.925
We want to be upfront with what it is we're taking, what it is we're doing,
00:26:02.169 --> 00:26:04.155
and what it is we may find.
00:26:04.980 --> 00:26:09.951
Ultimately, it's also helpful to ask, let's say in the case of someone who you're
00:26:09.951 --> 00:26:15.574
asking to take a DNA test, it's also very useful to ask them, are they,
00:26:16.108 --> 00:26:19.986
taking the test just to help you and they don't really want to know what you find?
00:26:20.328 --> 00:26:26.050
Or are they interested in knowing what you discover based on their information?
00:26:26.375 --> 00:26:31.021
Some people are happy to help, but they really don't want to know what it is
00:26:31.021 --> 00:26:32.533
that you're looking about.
00:26:32.875 --> 00:26:35.671
And other people are genuinely curious and would like to know.
00:26:35.671 --> 00:26:41.841
So it's good to ask them in advance so that there's no ambiguity,
00:26:41.841 --> 00:26:49.091
no problems later if you should discover something that wasn't necessarily what you expected to find.
00:26:50.191 --> 00:26:54.531
Yeah, just making sure that everything is open and on the table before anything
00:26:54.531 --> 00:27:00.781
gets started. So how do you see the future of genealogical bioinformatics evolving
00:27:00.955 --> 00:27:03.109
over the next decade? Sure.
00:27:03.815 --> 00:27:10.288
Basically, what I see, what I anticipate happening is that right now,
00:27:10.416 --> 00:27:14.746
our DNA testing providers have huge amounts of information,
00:27:15.402 --> 00:27:25.253
and our genetic genealogy tools basically allow us to usefully access the closest
00:27:25.253 --> 00:27:28.585
3% to 5% of our DNA matches.
00:27:29.044 --> 00:27:33.241
So, and most tools, again, if we're talking about the units of DNA matching,
00:27:33.241 --> 00:27:38.241
we talk about something called a centimorgan, and our best matches are going
00:27:38.241 --> 00:27:40.279
to be 40 centimorgans or greater.
00:27:40.703 --> 00:27:44.971
And those tend to be the top 3% to 5% of our DNA test matches,
00:27:44.971 --> 00:27:52.840
which means that 95% of our DNA matches aren't going to be very useful to most conventional tools.
00:27:53.380 --> 00:27:58.429
And what I anticipate happening is these bioinformatic tools.
00:27:59.208 --> 00:28:03.934
Can be used to basically can be used in conjunction with,
00:28:04.494 --> 00:28:07.594
and again, I hesitate to just put the word out there, but it can be used with
00:28:07.594 --> 00:28:10.235
artificial intelligence systems so that,
00:28:10.916 --> 00:28:18.104
our raw data, our DNA matches, our DNA sequences themselves can be inputs to
00:28:18.104 --> 00:28:21.442
systems that can then look for patterns, look for answers,
00:28:22.088 --> 00:28:24.560
look for correspondences. so that.
00:28:26.814 --> 00:28:31.784
We have questions and we can bring those questions to these tools,
00:28:31.784 --> 00:28:37.054
to these what we call agents or artificial intelligence agents to begin to do
00:28:37.054 --> 00:28:40.520
the work for us and work with our information.
00:28:41.002 --> 00:28:45.954
These agents can also guide us as to who we want to test or where we want to
00:28:45.954 --> 00:28:49.065
test or what information would be helpful or missing.
00:28:49.337 --> 00:28:55.054
And this is what I anticipate the future will bring for us basically the ability
00:28:55.054 --> 00:29:00.986
to use large amounts of data and for us to basically begin with questions,
00:29:01.591 --> 00:29:07.974
and have the computer systems assist us with our answers based on just looking
00:29:07.974 --> 00:29:11.294
through piles and piles of information that,
00:29:12.494 --> 00:29:17.164
could take us years to go through and that these computer systems will be able
00:29:17.164 --> 00:29:22.614
to both help us find answers to questions we have and also guide us as to where
00:29:22.614 --> 00:29:25.344
to look and instruct us really,
00:29:25.884 --> 00:29:28.614
what answers are there in the data we have.
00:29:28.614 --> 00:29:32.601
That if we have questions about our paternal lines,
00:29:33.591 --> 00:29:38.952
It may be that the DNA patterns and data actually don't have those answers.
00:29:38.952 --> 00:29:43.892
And in fact, the agents, our AI agents can then tell us, really,
00:29:43.892 --> 00:29:48.772
you should be looking at these maternal lines for your solutions because that's
00:29:48.772 --> 00:29:50.114
the information that we have.
00:29:51.442 --> 00:29:56.000
And ultimately, then at least we're using our time and our research efforts productively.
00:29:56.656 --> 00:30:01.802
Absolutely. And I guess for listeners who are curious but are intimidated by
00:30:01.802 --> 00:30:07.151
the silence, what would do you say to encourage them to explore this video?
00:30:08.409 --> 00:30:14.741
I would say basically there are right now on my website, Discovering New Ancestors,
00:30:15.095 --> 00:30:22.188
I have a bioinformatics page which provides descriptions of the tools that I've authored so far.
00:30:22.641 --> 00:30:27.663
And the basic descriptions describe what the tools can do.
00:30:28.278 --> 00:30:34.882
Let's say one tool may be able to give you a whole group of DNA matches who
00:30:34.882 --> 00:30:36.987
are connected to you by a certain ancestor.
00:30:37.672 --> 00:30:42.862
Another tool can do the kind of mystery match work that I described that I did
00:30:42.862 --> 00:30:44.207
with my great-great-great-grandmother.
00:30:44.671 --> 00:30:49.897
Another tool can take the matches of full sets of DNA matches of three people,
00:30:50.571 --> 00:30:53.787
and tell you whether they share a common ancestor or whether they don't.
00:30:54.756 --> 00:31:00.522
And these sorts of tools address specific problems. And if you basically read
00:31:00.522 --> 00:31:03.562
the description of one of these tools and said to yourself, well,
00:31:03.562 --> 00:31:07.962
this is a problem I have, then you can go a little deeper,
00:31:08.762 --> 00:31:13.362
and read more about, well, what do I need to do to use this tool and try to
00:31:13.362 --> 00:31:14.685
work for things from there.
00:31:15.630 --> 00:31:20.361
Also, at least knowing that these tools exist, even if you don't feel capable
00:31:20.361 --> 00:31:21.720
of doing the work yourself,
00:31:22.156 --> 00:31:27.321
at least you know that there's a tool which perhaps another researcher,
00:31:27.741 --> 00:31:31.163
or if you have a cousin who is, you know, very skilled at these sorts of things,
00:31:31.496 --> 00:31:36.362
at least you know that there is something to look at which might be able to solve the problem.
00:31:36.786 --> 00:31:40.490
Because again, not, I always like to sort of put it this way,
00:31:40.856 --> 00:31:44.745
you know, not everyone wants to do their own taxes. Yeah.
00:31:45.601 --> 00:31:47.131
But at least, you know, we know
00:31:47.131 --> 00:31:51.136
where the forms are and what we have to do and when it has to be done.
00:31:51.612 --> 00:31:56.141
And so maybe there's someone there who can help us. And these sorts of problems
00:31:56.141 --> 00:31:59.111
in family research are kind of similar.
00:31:59.111 --> 00:32:06.601
Not everyone is going to feel up to and able to really dig into all the little
00:32:06.941 --> 00:32:10.110
minutiae and problems that occur in family research.
00:32:10.360 --> 00:32:15.265
But at least knowing that the tool is there and that you have this information
00:32:15.654 --> 00:32:19.710
may be helpful in finding solutions with help from others.
00:32:20.268 --> 00:32:24.731
Absolutely. And what do you hope people ultimately gain, not just scientifically,
00:32:24.731 --> 00:32:28.823
but personally from engaging with genealogical bioinformatics?
00:32:29.467 --> 00:32:34.279
Well, my hope is, again, similar to what genetic genealogy has brought us,
00:32:34.563 --> 00:32:39.368
which has been basically answers and connections and information,
00:32:39.693 --> 00:32:43.901
either confirming what we already knew or providing us with insights.
00:32:44.818 --> 00:32:46.908
Into connections we weren't aware of,
00:32:47.757 --> 00:32:53.351
that really kind of bring us all a little closer to understanding our own heritage
00:32:53.351 --> 00:32:55.025
and our own place in the world.
00:32:56.029 --> 00:33:02.572
These are just new tools or enhanced tools for doing what researchers have been
00:33:02.572 --> 00:33:10.270
doing and wanting to do all of which is to make sense of difficult questions and to really sort of,
00:33:10.909 --> 00:33:14.232
bring information, to make sense of information,
00:33:15.032 --> 00:33:20.792
to bring, in this case, information science to bear on the biology that we've
00:33:20.792 --> 00:33:25.342
already been using in genetic genealogy, which itself is just bringing more
00:33:25.342 --> 00:33:30.022
evidence to what researchers have already been working on for decades.
00:33:30.625 --> 00:33:34.427
Yeah. And, you know, Arun, as we come to a close with this episode,
00:33:34.654 --> 00:33:37.942
where can listeners learn more about your work and, of course,
00:33:37.942 --> 00:33:39.629
continue exploring this field?
00:33:40.273 --> 00:33:44.459
Well, I have a website, which is discoveringnewancestors.com.
00:33:45.389 --> 00:33:51.044
There's also a Facebook group I started, Genealogical Bioinformatics,
00:33:51.398 --> 00:33:57.033
which I started basically as a way of sort of getting people online able to
00:33:57.033 --> 00:33:59.689
discuss different issues and problems.
00:34:00.182 --> 00:34:04.041
There's, I think, a hundred or maybe so members of the group now,
00:34:04.384 --> 00:34:11.121
and it's not terribly active, but it's something that I hope to engage with more in the future.
00:34:11.579 --> 00:34:18.903
Something I recently began is a project to document and basically prepare educational
00:34:18.903 --> 00:34:20.738
materials about the subject.
00:34:21.110 --> 00:34:26.993
I was awarded a research grant by the American Society of Genealogists to prepare
00:34:26.993 --> 00:34:30.750
educational materials on genealogical bioinformatics.
00:34:30.982 --> 00:34:37.223
And so I hope to preview some of those materials in the Facebook group to get,
00:34:37.863 --> 00:34:40.124
feedback from interested people there also.
00:34:40.565 --> 00:34:44.463
So I would certainly start with my website, DiscoveringNewAncestors.com,
00:34:44.783 --> 00:34:48.422
and there are links to the Facebook group and other things there.
00:34:49.096 --> 00:34:52.823
Perfect. Thank you so much. I'll make sure to put that in my show notes so people
00:34:52.823 --> 00:34:55.308
can connect with that. But I just wanted to thank you so much,
00:34:55.592 --> 00:35:00.033
Arun, for coming on to the podcast, talking about this, and so people can be
00:35:00.033 --> 00:35:01.983
able to explore. So thank you so much.
00:35:02.308 --> 00:35:05.194
Oh, thank you for having me. It's been my pleasure. No, no problem.
00:35:06.796 --> 00:35:11.973
Hope you enjoyed this episode and if you did please make sure to like follow
00:35:11.973 --> 00:35:18.377
subscribe and write a review for the episode wherever you listen to your podcast thank you.