Genetics For The Rest Of Us: Mail-to-order genomics--worthwhile?

NPR recently ran an article discussing why mail-to-order genomic analysis may not be all it’s cracked up to be, and I thought it would be worthwhile to continue that discussion. Companies offering this service are advertising promises that I feel they are often unable to deliver, and I think people should know why scientists like me should be wary of these promises.

In this post, I’ll explain what mail-to-order genomic analysis is, what the genome is made up of, why the genome can be difficult to interpret, and why the genome doesn’t provide the whole picture of human health.

What is mail-to-order genomic analysis?

Many of you have probably seen ads for these services. In a nutshell, companies sell kits so you can learn more about your genome. You can order these kits online, and a company mails you a tube to spit in so you can send it back for analysis. A few weeks later, they give you the results.

One such company, 23andMe, first started selling their kits to customers in 2007, the company was primarily providing novelty genetic information¹. For example, they would look your genome for variants associated with dry earwax or whether your urine was likely to smell particularly bad after eating asparagus.

A few years later, 23andMe got in trouble with the U.S. Food and Drug Administration (FDA), the federal agency that oversees food safety and pharmaceutical drug approval, when they started providing health information in their analysis reports^2,3.

What the FDA was concerned about was that patients would use 23andMe’s genome analysis to make their own health decisions instead of talking to a physician^2,3. This is a problem because genome analysis is relatively new compared to a lot of biology, so scientists and physicians are still don’t understand how variants relate to many diseases.

What exactly makes up the genome?

The human genome is over 3 billion nucleotides (letters that make up the genome, abbreviated as A, C, T, and G) long and split into 23 separate chunks called chromosomes.

Genes in the genome provide instructions for individual components of the cell. Simply put, the genome is the entire instruction set for making an organism. The genome is housed in the cell’s nucleus, the “command center of the cell,” and the cell is constantly referring to it in order to do things it needs.

To relay the necessary information in the genome to the rest of the cell, a gene is transcribed into ribonucleic acid (RNA) and exported from the nucleus. That RNA can encode a message for a protein, an extremely large molecule that performs a given function in the cell, to be made. Outside the nucleus, ribosomes translate the RNA message into a protein.

For decades, we thought each gene made a single protein and RNA was nothing more than a message. We thought that if we sequenced the human genome and found what all the genes were, we’d have a mostly complete understanding of the cell and what makes us human.

Why can the genome be difficult to interpret?

When we first sequenced the human genome, we were surprised to find only 3% of it was made of genes⁴. Moreover, we were expecting to find around 100,000 genes, and we found only around 26,000-30,000. Now, some scientists are the count is even lower, maybe even less than 20,000⁵.

What this basically meant was that 97% of the genome was a mystery.

Around the same time, other biologists were figuring out that RNA did a lot more than we thought, maybe just as many different things as proteins could. Looking at the genome again, we found well over 9,000 genes that didn’t make proteins and instead made RNA with unknown function⁶.

Also, we started learning that each gene could be read in multiple ways to produce slightly different proteins that could have very different functions in the cell⁷. So, the idea that one gene was supposed to make only one protein, was also wrong.

So, saying the genome is complicated is a major understatement.

Why is the genome not the whole picture?

As it turns out, the DNA making up the genome itself can be modified to affect how the cell reads the genome⁸ in response to environmental changes (like if rat mother doesn’t lick her babies enough, certain genes don’t get turned on properly; you can pretend to be a rat mother here). Traditional genome sequencing doesn’t catch these modifications.

We also found out that the way the genome was packed into the nucleus was important. Take progeria, for example, a disease that causes rapid aging (like in the movie Jack with Robin Williams). A few years ago, we found an error in tethering the genome to the nucleus can cause a form of the disease⁹. We’re still trying to figure out how genome packing can affect health and disease, but genome sequencing doesn’t really help us understand it.

Alone, the genome sequence is just one piece of a giant puzzle, which is why the got in trouble with the FDA. It’s misleading to say we can know a lot about your future health by looking at the genome. 23andMe is still a successful company, but they’ve had to scale back on what they way they can and can’t do¹⁰.

So is mail-to-order genomics worthwhile?

Companies like 23andMe certainly told customers genomic analysis was worthwhile.

Doctors already routinely screen for errors in single genes, so genetic information does have real, practical uses. Medicine is certainly heading in the direction of whole genome analysis. The National Institutes of Health (NIH) is pouring money into it, and at some point, genome sequencing will be an everyday part of our lives. But that time is a long way off, I think.

So is mail-to-order genome analysis worthwhile? Not yet, in my opinion.

References

Genetics For The Rest Of Us

Monday, July 13, 2015

Mail-to-order genomics--worthwhile?

No comments:

Post a Comment