Rainbow of variation: A schematic of worldwide human genetic variation, with colors representing different genetic types. The figure illustrates the great amount of genetic variation in Africa.

Credit: Illustration by Martin Soave/University of Michigan

Related articles

• Human genome surprisingly diverse
• First individual person's genome decoded
• Stomach ulcers map human origins
• Gene makes Africans more susceptible to HIV
• First Europeans came from Asia

PARIS: Diving deep into the human genome, scientists have drawn up the most detailed maps to date of our evolutionary past. Their findings are detailed in two studies published this week in the British journal Nature.

One paper reveals that human genetic diversity decreases the further one gets from Africa, the cradle of humanity.

People of African descent are more varied genetically than people of the Middle East, who are in turn more diverse than either Asians or Europeans, the study found. By the time Homo sapiens migrated to the Americas across the Bering Straits, diversity had dwindled even further.

Damaging mutations

The other investigation shows that Americans of European descent have more potentially damaging mutations in their DNA than African-Americans, a finding that settles a long-standing debate.

It is now clear, the researchers say, that all persons of European descent, and not just isolated geographic groups, experienced a "genetic bottleneck" – probably between 30,000 and 100,000 years ago – as a small, founding population moved into present-day Europe.

As a result, the gene pool in Europe was restricted, and possibly harmful mutations in DNA were handed down over the generations instead of being flushed out of the genome through the evolutionary process of natural selection.

Both papers have important implications for understanding the genetic origin of disease and why some populations – associated with ethnic groups, geographic locations or both – seem more at risk from certain disorders than others.

Huge data sets

In the first study, a team of international researchers led by Noah Rosenberg of the University of Michigan and Andrew Singleton at the National Institutes of Health (NIH) in Bethesda, Maryland analysed DNA from 485 people in 29 different populations around the world.

They identified and compared 500,000 DNA markers – slight variations in genetic code in a string of three billion pairs of chemical compounds – in the human genome.

"Now that we have the technology to look at thousands, or even hundreds of thousands, of genetic markers, we can infer population relationships and ancient migrations at a finer level of resolution than as previously been possible," said Rosenberg.

In the second study, Carlos Bustamante of Cornell University in New York state and his team sequenced more than 10,000 genes – nearly half the total human genome – in 20 Americans of European ancestry, and 15 of African heritage.

They found 40,000 locations with DNA variations in at least one individual, and just under half of those were capable of altering amino acids, meaning they could have potentially far-reaching effects on body chemistry.

The proportion of a subset of "probably damaging" variations in DNA base pairs, called single-nucleotide polymorphisms (SNPs), was 12.1 per cent for African-Americans and 15.9 per cent for Americans of European descent.

Tracking bottlenecks

Future investigations should reveal whether other populations went through a similar "bottleneck" as they moved further from Africa, leaving deleterious variants that remain in the DNA today. Bustamante said his findings opened up new paths for understanding inherited disease but could not be used to predict health outcomes for individuals alive today.

Genomics is such a young science that far more work is needed to comprehend the genetic sources of disease and to map individual genetic codes, he said. "The results of this study suggest that everybody carries in their genome hundreds of mutations that are disrupting protein structure."

"But in order to correlate them with disease you need to compare sequenced proteins from individuals with and without the disease in order to correlate them with specific mutations."