|
Evolutionary
history shows that human populations likely originated in Africa. How
they migrated from this continent, however, has been a source of debate
and conjecture for many years. Until now, the logical explanation was
that people left by way of Egypt, the Sinai Peninsula, and the into the
Ferticle Crescent, where they then spread west to Europe and east to
Asia.
New
findings from a joint effort by National Geographic and IBM Research,
however, suggest that modern humans migrated from Africa via a southern
route through Arabia instead. Leaders
of the Genographic Project, the most extensive survey of human
population genetic data to date, suggests where they went next,
announced their findings today at a conference at the National
Geographic Society in Washington, D.C.
In
2005, geneticist and National Geographic Society Explorer-in-Residence
Spencer Wells, who had already embarked on DNA analysis projects to
solve questions about the origins of modern humans, suggested that
better solutions for analyzing patterns in DNA might offer valuable
clues to early human migration patterns. His idea attracted the
attention of Ajay Royyuru, head of the Computational Biology Center at
IBM Research. It was a departure from his research efforts in structural
biology, but he found Wales’ thoughts compelling and helped lead
development of a new analytical method that traces the relationship
between genetic sequences from patterns of recombination—the process by
which molecules of DNA are broken up and recombine to form new pairs.
By
looking at similarities in patterns of DNA recombination that have been
passed on and in disparate populations and inferring relationships
between them, scientists have confirmed that African populations are the
most diverse on Earth, and that the diversity of lineages outside of
Africa is a subset of that found on the continent. Divergence in the
common genetic history between populations showed that Eurasian groups
were more similar to populations from southern India, than they were to
those in Africa. This suggests that people from Africa crossed the
Bab-el-Mandeb Strait in Arabia before embarking on any movement heading
north.
According
to Royyuru, the most daunting task for researchers, and the most
crucial aspect of the project, was successfully analyzing of the
recombination portion of the genome, the vast segement of the genome
that is passed from generation to generation.
Genetic
information is not passed directly from generation without a
reshuffling of the genetic elements, producing a unique outcome for each
offspring. In the past, geneticists had shied away from analyzing that
99% of the genome because that was difficult to figure out which portion
of the genome came from which parent.
“Almost
99% of the genetic makeup of an individual are layers of genetic
imprints of the individual’s many lineages. Our challenge was whether it
was even feasible to tease apart these lineages to understand the
commonalities,” says IBM researcher Laxmi Parida, who defined the new
computational approach in a study published in Molecular Biology and
Evolution.
“This
was a computational bottleneck that sat there for quite a long time,”
says Royyuru. “Everybody thought it was an intractable problem.”
The
breakthrough was to develop an algorithm—one that Royyuru says
represents a new approach, mathematically—that would analyze that
portion of the genome and infer recombinant history in a way that is
sufficient enough for researchers to obtain useful information. The
major limitation was that recombinatorial data from just one generation
could be used.
“Think
of genes lined up on a chromosome like a deck of cards. A chromosome
from the father, a chromosome from the mother—these are reshuffled to
produce another deck of cards,” says Royyuru.
To
figure out what elements came together, and the type of shuffling that
occurred, researchers can only use the genomes of the present
generation. Theoretically, says Royyuru, reconstructing the entire
pattern of recombination through generations is impossible. But enough
portions can be parsed so that researchers at least get a glimpse of the
pattern.
“I
would categorize this as a new approach. Previous analyses have focused
on the presence or absence of individual variations in the genome. They
haven’t looked at what you can infer,” he says.
Ajay Royyuru, head of the Computational Biology Center at IBM Research. |
The
inferences produce a result that is entirely unlike the familiar family
tree. The graph of recombination, which traces across lineages as well
as forward through time, looks more like a web than an expanding chart
of branches.
The
computational solution was first tested against a synthetic genome. The
results convinced project leaders to begin collecting samples. In
addition to the 11 principal investigators, up to 60 others were
involved in a massive sample collection project that was tasked to
collect 100,000 voluntary samples and 100,000 field samples. Royyuru
says the project has produced more than 450,000 public samples and more
than 75,000 field samples. The next step was an effort to get
genotypting information, for which the team looked specifically at the X
chromosome in males.
“This
is given to us by other mothers. In the mother, as the egg is produced,
a recombination of the two X chromosomes is created. The history of the
X chromosome is carried in the chromosomal history, and we are able to
see pattern of global migration that is uniquely inferred from this
global history.”
The
analysis gave the IBM and National Geographic researchers the clues
they sought as to the most recent major migration of humans from Africa,
which may have occurred about 35,000 years ago. The huge sample size,
coupled with about 1,000 samples from a diverse geographic area that
were specifically analyzed for their X chromosomes, gave researchers the
confidence to assert that migration most likely occurred by leaving the
horn of Africa, entering southern Arabia, and from there spreading into
India.
Royyuru
says that the Genographic Project’s findings will be tested further in
collaboration with other research groups. Archaeological evidence will
be analyzed to back up the findings, and he hopes to apply this method
to the entire genome and the other 22 chromosomes.
“When
we started, our goal was to bring science expeditions into the modern
era to further a deeper understanding of human roots and diversity. With
evidence that the genetic diversity in southern India is closer to
Africa than that of Europe, this suggests that other fields of research
such as archaeology and anthropology should look for additional evidence
on the migration route of early humans to further explore this theory,”
says Royyuru.
