Breakthrough of 2007: Human Genetic Variation

Science Magazine Declares that the DNA Differences Between People is the Breakthrough of 2007

January 24, 2008

Each year Science magazine proclaims some scientific topic to be the breakthrough of the year. A couple of years ago it was evolution. This year it is human genetic variation.

Human genetic variation just means the DNA differences between any two people. Scientists found a lot of new differences in people's DNA last year. And they have come to realize that a previously overlooked difference called copy number variation (CNV) is much more important than anyone realized.

In 2007, scientists used these differences to discover genes that contribute to complicated disease like diabetes, multiple sclerosis, Alzheimer's, and many others. They also used these differences to figure out why people look different from each other.

This was also the year that four new companies were launched that let people find out about their DNA in a big way. There has been a lot of discussion about whether this is a good thing, especially given how little scientists actually know right now about human DNA.

More SNPs, More Studies, More Disease Genes

The kind of DNA difference that most of us are familiar with are SNPs or single nucleotide polymorphisms. These are differences in the letters of DNA.

DNA is "written" in a chemical language that has 4 different letters (or nucleotides) called A, G, C, and T. Each person has his or her 6.4 billion letters of DNA all arranged in a certain order. Someone else has around 6 million differences in these letters scattered throughout his or her DNA. This is a big part of what makes people different from one another.

The cost of figuring out a person's DNA continues to get cheaper and easier. This has allowed scientists to identify many more of the DNA differences between people. Knowing more of these differences has meant that scientists can fine tune studies looking for genes involved in diseases. And so begin to get a handle on a patient's risk for getting a disease.

Scientists can put 500,000 or even a million of these SNPs into a single genetic test using microarrays. They can then use these tests to compare the DNA of people with a disease to the DNA of people without the disease. Doing this allows them to figure out which differences are important risk factors for a particular disease.

These tests can be used to find genes that have small effects because they are now cheap (and good) enough to be used on thousands of people in a single study. This matters because most common diseases come about because of the influence of the small effects of many genes (and the environment). Last year scientists found genes that contribute to the following diseases:

  • Heart disease
  • Breast cancer
  • Restless leg syndrome
  • Atrial Fibrillation
  • Glaucoma
  • ALS (Lou Gehrig's disease)
  • Multiple sclerosis
  • Rheumatoid arthritis
  • Colorectal cancer
  • Ankylosing spondylitis
  • Autoimmune diseases

Scientists have also used these tests to identify genes involved in how we look. For example, they identified a gene that partly explains why some people are taller than others. And genes involved in skin, hair, and eye color too.

Listed below are links to some of the articles we've written on this subject.

New Way to be Different: Copy Number Variation

Besides SNPs, scientists have realized that another kind of difference is much more important than they thought. This difference is called CNV or copy number variation.

Scientists have known about CNVs for a long time but they hadn't realized how common or important they were until this year. Instead of a difference at a single letter like with SNPs, a CNV has changes in the DNA that involve many letters.

For example, sometimes a few letters are repeated or go missing. This will affect how a single gene works. Or sometimes up to a few million bases are repeated or go missing. This can lead to situations where people have extra copies of certain genes.

Scientists found that these insertions and deletions were much more common than they thought. In fact, they are common enough that their discovery has prompted scientists to rethink how similar any two people are.

Based on just SNPs, scientists have always said that any two people are 99.9% the same. But when you factor in CNVs, the similarity drops to 99.5%. Still pretty similar but more different than before.

And these CNVs contribute to disease too. For example, repeats of the CCL3L1 gene increase someone's resistance to HIV and so to AIDS. CNVs can also explain how autism sometimes appears in a family tree.

CNVs may be important in evolution as well. For example, people whose ancestors ate a lot of starch have extra copies of a certain starch processing gene called the amylase gene. It has been proposed that this is the result of natural selection. Basically, people in these cultures who could digest starch better were more successful. And so having extra amylase genes became more common.

The study of CNVs is still at a very early stage. Undoubtedly as scientists dig deeper, they'll find that CNVs are involved in many more diseases. And that they're an integral part of who we are.

Listed below are links to some of the articles we've written on this subject.

More Information


Scientists can look at 500,000
DNA differences at once
using microaarays.

Arrays like this can
find new CNVs.
Image courtesy of Agilent
Technologies, Inc.

Personal Genomics Takes Off

In 2007, four companies that promise a detailed look at your DNA opened for business. For 1000 dollars or so, 23andMe and deCODEme look for 500,000 different SNPs in someone's DNA (similar to the tests used to look for genes related to disease). For 2500 dollars, Navigenics will scan a client's DNA and give him or her a report on 20 or so different diseases. And Knome will tell a client what all 6.4 billion letters of his or her DNA look like for 350,000 dollars.

A question many people in the scientific community are asking is, "Are any of these tests worth the money?" The answer depends on what a client wants from them.

If the test is done for fun and a client has the money, then it can be worth it. The client can find out a bit about his or her ancestry, find out if he or she carries the gene for red hair, and other recreational genomic pursuits. If this is a client's idea of fun and he or she has the money to spend, then it is worth it.

These tests can be less useful if someone wants to figure out his or her risks for getting a disease. And they can be even worse than useless for some people.

A hypochondriac would go nuts with these results. The hypochondriac would find out that a certain DNA sequence he or she has leads to a 2.3 fold higher chance of developing Type 2 diabetes. And another change predisposes him or her to restless leg syndrome. Now repeat this kind of thing hundreds or thousands of time.

Now factor in that lots of these differences will turn out not to be important. Because the field is so young that a lot of the findings are preliminary and will change as more studies are done.

And what if someone finds they will get some awful disease in the future? 23andMe and deCODEme don't offer genetic counseling so the person is on his or her own to find counseling (if they even know what it is).

Thus begins the brave new world of personal genomics. Right now scientists don't know a lot about our DNA. And what they do "know" changes or is found out to be just plain wrong. (A little bit of knowledge is a dangerous thing!)

So tests like these should be taken with a grain of salt. People should use them for entertainment purposes. And think twice about using them to plan a healthy life.

How much do you want
to know about your DNA?