Using DNA to Explore Ancestry (Panel)


Mark Shriver:
All right, so we’ve got time for a few questions for our panelists, so, please, we encourage
the audience to step up. Linda Heywood:
Yes, thanks very much for this very informative and very scientifically up-to-date presentation.
I wanted to direct my question to Rick Kittles, because I saw that you had on your list for
West Africa Bamileke from the Cameroon, and, in fact, we’ve been saying that, in fact,
the Cameroon did not provide slaves to — enslaved Africans, so we have to see the Bamileke Cameroon
as a sort of stand-in, for now, for the Angolan population. Because that’s where — so I just
wanted to found out if your database has been updated, or if there’s — because a lot of
people come to tell me, “Well, I’m from the Cameroon,” and I said, “You can’t be, have
ancestry when there were no slaves at the time between, you know, 1619 and 1860.” The
Cameroon was not that theatre for the — now some of them could be Igbo, because we know
that Cameroon — but you have to take into account that some of those, that Bantu population,
passed through certain genetic kind of a transmutation, as they’re made the way down on the Eastern
Bantu and the Western. So I just wanted to make sure that more — that more — no, I
have to say that — Rick Kittles:
So — right. Okay. Linda Heywood:
— that we do the research and collect data for Angola. Rick Kittles:
Right. No, you’re absolutely right. And I liked exactly what you said just now. We have
to do the research. So based on what you guys have been saying, in terms of the historical
research, the historical evidence that Cameroon, present-day Cameroon region, right, did not
contribute any of the enslaved Africans that were traded. Now that is still up — I mean,
that’s not a definitive. There are folks who do not believe that, and if — particular,
if you look on the coast — I just came from Cameroon last December. And I went to a fort
along the coast of Cameroon, and where enslaved Africans were housed, where there was a point
of no return. So there are, you know, historical remnants. I mean — Linda Heywood:
[inaudible] That’s what I’m saying, I’m not saying to the Americans. Rick Kittles:
Well, we know also that — let’s say they went to the Caribbean, that there was an enormous
amount of selling and buying of enslaved Africans in the Caribbean to the Americas, too. So
they may not have directly come here, but they could have come here via the Caribbean.
But it doesn’t matter. Let’s say you’re right, okay? We see Cameroon as a transitional area
anyway, because we call it — I mean, it’s been called the Bight of Biafra, and you look
at the genetics of the Igbo communities, and some of the Cameroon groups like the Bamileke,
they’re very, very similar. That border is a border that was given to them, you know,
by European — you know, when was it, after World War I or II? You know, all those boundaries
were set up. That area consisted of families that actually were split after that boundary
was set up. And then, of course, the Biafra war, there was back and forth, the boundary
went this way, that way, but the bottom line is that’s a transition area. So I see it as
a transition of what we see in Nigeria and a transition, a good point, of Gabon and Angola,
okay? So when we do our matches, we say where the match is to a present-day country. That
doesn’t mean that that was a present-day position where, or a state, where enslaved Africans
came from, where their ancestor came from during slavery. You see what I’m saying? We’re
matching to a present-day population. Mark Shriver:
Great, thanks, Rick. Next question? Female Speaker:
Hello. This question is for Dr. Mountain. I was wondering if you could share a little
information about your Roots in the Future initiative that was launched in 2011, if you
could share a little bit about the successes as well as the challenges that 23andMe has
faced. Joanna Mountain:
Yes. So — is this on? Okay, so, in 2011, 23andMe launched a program called Roots in
the Future, and the reason for launching that program was because we were conducting research
into the links between disease risk and genetics by inviting our customers to take surveys
about their health and disease status. So what was happening, though, is it turns out
that all of the studies we were doing were relevant only to people with Northern European
ancestry, because those — most of our customers have ancestry tracing primarily to Northern
Europe. And so it was very frustrating for us to see that we were, just like the rest
of the research genetics community, were kind of being restricted to, you know, just really
a subset of the world in terms of our research, our genome-wide association study. So we decided
to launch this project to enroll 10,000 African-Americans into the 23andMe service and invite them to
take part in research. So that project has been incredibly successful,
as I think I put the number up, we have well over 10,000 African-Americans have signed
up and joined 23andMe, and we have begun conducting the research to look at connections between
health and genetics within that particular group. It’s obviously a very heterogeneous
group. And we’re also exploring, you know, other aspects of the customer base through
surveys, and also, so the genetic data by including looking at ancestry, as I started
to show here. So the project has been very successful, very high survey response rate
among the group, and so engagement in the project, and we’ve been looking to start writing
up some of the research results, and focus on the disease research. Mark Shriver:
Thanks, Joanna. Charles? Charles Rotimi:
Yeah, again, this is question to all members of the ancestry panel, and also to the audience.
I just wonder, morally, is it right for us to be charging African-Americans to trace
their ancestry using this kind of information? Given, using Rick’s word, that they were “kidnapped,”
were forcibly removed from the environment that it actually has a link to? So why are
we now charging for them to be able to trace that information back? Should there be a fund
set aside that any African-American who wants to use this kind of information should be
able to tap into their funds and to derive this kind of information? Rick Kittles:
Yeah, I agree with you, Charles, there should be a fund, and you should be the first to
contribute. [laughter] Charles Rotimi:
It’s really — to me, it looks like double jeopardy, in a sense. Rick Kittles:
No, what you raised is a really valid point, I mean, and it’s been part of the discussion
since, you know, we started doing this. And so, you know, the issue is, “Should African-Americans
who are descendants of enslaved Africans have to pay for what was taken from them?” Mark Shriver:
Yeah, and, you know, this is — I mean, the price has come way down, and, in fact, that
issue we were talking about was pro bono, right, Joanna? Joanna Mountain:That’s right. That particular
project — Mark Shriver:
Yes, those 10,000 testes were free — Joanna Mountain:
— was — yeah, were not — yes. Mark Shriver:
— to people that enrolled early enough. Joanna Mountain:
That’s right. Mark Shriver:
Okay, next? Female Speaker:
My question’s really for the panel, it just seems like such a wonderful opportunity to
have representatives here from different organizations that are conducting these ancestry tests.
And I think one of the challenges we heard from Dr. Tishkoff this morning, and that continues
to be a problem, is obviously how to get accurate test results. And I wonder, to what extend
do you think it’s a responsibility or desirable aim for your companies to have similar test
results for people, and are there steps that you’re taking to get consistency among organizations,
and if so, how can you fit this with your commercial objectives? Rick Kittles:
Well, first — can I say something? You shouldn’t expect that you can get the same results from
different companies. I mean, the reference databases are different. What the companies
can do, though, is work together to some extent to come up with a set of standards that you
could expect consistency on. But, you know, certain tests for, you know, certain companies,
have unique tests, and that’s what makes the company unique. But I do think that there’s
room for the companies to work together to come up with a set of standards. Mark Shriver:
Okay, so we’re over time, but Dr. Blakely, just 20 seconds? Michael Blakey:
Thanks, I have a question of Dr. Mountain. I was interested in Roy, and it seems to me
that what you’ve found with Roy is you weren’t able to identify, had you just the DNA data
of a village, or a nation, or an ethnicity. You’ve essentially got sort of race admixture
for him. But what interests — but what I’m really asking is, do you have more Roys? Because
at this point, it’s anecdotal. Do you have a sample of 30 Roys so we can see when it
works and doesn’t? Joanna Mountain:So there — what I think what
we’re hearing here are a couple of questions about how to be — have more confidence in
the results so that customers can have more confidence in the results. And so one way
is to have 30 Roys who know their, you know, history, and then we see if we recover that
through DNA. And — but what we — the other way is to simulate individuals, simulate Roys,
or stimulate, and so that we do sometimes do something like that. We also have this
approach where we look at the precision and recall, where we ask — which is a little
bit different — but we just ask, “How often, if you do have a segment that does — should
be identified as coming from one region, how often does it actually — the DNA show that?”
And we run our — we keep tweaking our algorithms until that’s really high. And then we actually
allow our customers to choose how their level — the level of accuracy that they want to
look at. And so that’s another thing that we do in that direction. But finding lots of people who know — have
— who will kind of be gold standards, who really know the truth, so to speak, then we
can see if we can recover that. That’s the challenge. And it’s something we’ve been discussing
earlier this week. If it’s someone who would send to each of the companies these kind of
gold standard people and samples, and say, can you — “Tell us what your best estimate,
and we’ll see if — you know, which companies get it close to what we picked.” Mark Shriver:
Yeah, so send us your samples, everybody who has good genealogies. Joanna Mountain:
Yeah, yeah. Mark Shriver:
And we’ll have to stop here. Want to thank the panelists, very informative. Joanna Mountain:
Thank you, Mark. [applause]

Tags:, , ,
One Comment

Add a Comment

Your email address will not be published. Required fields are marked *