En 2008, la revue Science a publié l'étude génomique la plus complète jamais effectuée. Cette étude compare 650 000 nucléotides chez 938 individus appartenant à 51 ethnies. Les nombreux généticiens qui ont participé à ce travail ont conclu de leurs travaux qu'il existait sept groupes biologiques parmi les humains : les Africains subsahariens, les Européens, les habitants du Moyen-Orient, ceux d'Asie centrale et d'Asie du Sud, les Asiatiques de l'Est, les Océaniens et les Amérindiens.
Interviewer - Stewart Wills
The worldwide genetic diversity of humans is a matter of both basic biology and how
humans have moved across the globe over time. DNA studies can provide insights into
both dimensions. Now, a team of researchers from Stanford University has put together
the largest genetic exploration of human diversity yet. I spoke with two of the study's
authors, Richard Myers and Gregory Barsh, about the work and its significance. Dr.
Myers began with a quick description of what the work was all about.
Interviewee – Richard Myers
What we did was, we measured genetic variation in more than a half a million places
around the human genome, in almost a thousand people whose DNA had been collected
by geneticists over a several-decade period, led by Luca Cavalli-Sforza here at Stanford.
And this culminated in the formation of a group called the Human Genome Diversity
Project – these are the samples that we analyzed. These people are from 51 regions,
really spread around the world; although there are regions that are not well covered, it
shows a, quite a bit of diversity from around the whole planet. What we did was that, by
measuring the genetic variation of each of these loci – which is now made possible by
very recent technology, to do this fast and cheap and accurately – we were able to get a
picture of the genetic differences and similarities between people in these populations
from around the world, at much, much, much greater detail than had been measured
before.
Interviewer - Stewart Wills
So you mentioned the Human Genome Diversity Panel that you, that you used – what
made that population a particularly good vehicle for studying this kind of thing?
Interviewee - Richard Myers
These 51 groups, regions of the world, were places where geneticists and anthropologists
worked together over several decades collecting DNA samples from individuals who
came from those regions. And what’s particularly useful about this is that they are from
all over the planet, not just in one continent or one region.
Interviewee – Gregory Barsh
Right – what really makes the panel unique and different from what’s been done before is
the large number of individuals and their diversity in terms of the different regions
around the planet where they’re from. I think it’s worth comparing that to another very
large group of individuals that has formed the basis for many disease studies, genetic
disease studies; it’s called the HapMap. And the HapMap is based on four populations,
two of whom are very closely related. And what’s different about the HGDP sample is
it’s based on 51 populations, who when we into it, we really didn’t know how they were
related to one another.
Interviewer - Stewart Wills
So this is potentially a much, a much bigger sort of window into human diversity
generally.
Interviewee – Gregory Barsh
Absolutely.
Interviewer - Stewart Wills
Well let’s drill down now a little bit more into what your group specifically did with
these samples. What, how did you go about studying them to kind of, to kind of address
this problem?
Interviewee - Richard Myers
What we did was we took the DNA – and by “we” I mean 2 senior scientists in our group
at the Stanford Human Genome Center, Dr. Jun Li and Dr. Devin Absher – and what,
what Jun and Devon did was they took the DNA, and by using a genotyping system, we
measured DNA sequence variation at 650,000 loci, or 650,000 regions of the genome,
that were known to vary in at least the HapMap populations. These are all singlenucleotide
polymorphisms, or single bases that differ – they’ll have either an A or a G for
instance at a particular position.
Interviewer - Stewart Wills
So you were starting out with what had already been found about specific parts of the
genome where these single nucleotide variations were common, and then you were, you
were looking at those specific areas across this larger, more diverse data set.
Interviewee - Richard Myers
That’s exactly right. And so then – we call that genotyping, of course – and so we
genotyped these individuals – about 1,000 individuals. We actually chose the thousand
of that set of the Human Genome Diversity Panel that were from the 51 populations, none
of whom were related to each other, at least not related to each other in very recent
history. Obviously we’re all related to each other, and that’s one of the things you learn
from this type of study. And so, we collected those data, and that actually only took a
several-month period for us to. But then we spent a lot more time analyzing the data.
Interviewer - Stewart Wills
What are some of the more interesting things that you have found about human genetic
diversity and populations in looking at these data?
Interviewee – Gregory Barsh
Well, I think that one of the most interesting things is the realization that we are, in fact,
all related to one another. And looking at the way in which these 650,000 places in the
genome vary, tells us the ways in which we’re related. One of the striking things to
emerge from the analysis is that many people in populations that we would have thought
of as relatively homogenous, actually have ancestors or ancestry from multiple different
continents, and that’s particularly apparent in the Middle East. And in retrospect that
makes sense because the Middle East, of course, has served as a bridge as early humans
migrated out of Africa into Europe. And that’s reflected now when we look at
populations that are in the Middle East, and we see well where are their ancestors from,
in an agnostic way, that is – you know, let their DNA sequence tell us where their
ancestors are from. And we see ancestry in Middle Eastern populations from both Africa
and Europe. Another very nice example is the relationship between the Yakut population
and Native American populations. And that, of course, makes lots of sense when we
think about the origin of Native American populations. I think it’s also worth stating
explicitly that the study adds to the now I would say overwhelming support that we have
for the notion of where humans are from initially – that is, early humans really first
appeared in Africa and then populated the rest of the world as they migrated out of
Africa.
Interviewer - Stewart Wills
Well, we’ve talked a bit about population – does this help us also in genomic medicine?
Interviewee - Richard Myers
Having the ability to understand the genetic variation within groups will help us in
genetic disease studies. One of the biggest problems, and I’ll let Greg elaborate on this,
but one of the biggest problems is that when you don’t take population or geographic
origin into account in a, in a large genetic study for studying something like heart disease,
one of the complex traits for instance, you end up confounding the study such that you
don’t actually get real signals.
Interviewee – Gregory Barsh
It’s clear that many important common human diseases vary in their incidence among
different populations. And sometimes that clearly is due to culture – for example, the
incidence of obesity in western society. You know, other times it’s probably due to
genetic differences between populations. And many times it’s probably due to a
combination of environment and ancestry. And figuring out what the ancestry is for any
human on the planet is really a prerequisite to deconvoluting the interrelationship
between ancestry and environment in understanding common human diseases.
Interviewer - Stewart Wills
Gregory Barsh and Richard Myers are the authors, with nine colleagues, of a new study
that provides the most detailed description to date of human genetic diversity. The paper
appears in the February 22nd issue of Science.
http://www.sciencemag.org/content/319/5866/1100.short
http://www.sciencemag.org/content/suppl/2008/02/21/319.5866.1118b.DC1/SciencePodcast_080222.pdf
A very interesting paper:
http://ann.sagepub.com/content/661/1/65.abstract
Encore un super papier de Risch sur les clusters:
http://www.genetics.org/content/200/4/1285.abstract#corresp-1
Mais comment arrivent ils à séparer à différencier les européens des asiatiques, sinon par des clusters de gènes partagés.
http://www.sciencemag.org/content/346/6213/1113.abstract
Le débat sur les réseaux sociaux: