Waiting for the Genomics Payoff
David Barker, chief scientist at Illumina, explains how his company's technology works -- and why investors need to be patient

Scientists often say biotechnology has entered the "postgenome era." Translation: Now that scientists have identified the estimated 30,000 to 60,000 genes in the cells of the human body, it's time for them to make sense of what they've discovered.

David L. Barker, the chief scientific officer at Illumina (ILMN ), helped create -- and is now trying to improve -- the DNA-screening technology that's being used to do billions of necessary tests on the tiny variations in genes that are thought to hold the secrets to disease. Illumina, based in San Diego, is one of dozens of companies that have started to dig into the answers contained in the genome's raw data.

The process is time-consuming, and investors are running short on patience, driving down the stocks of these companies. Illumina now trades at around $4 a share, down from $12 at the beginning of the year. On Oct. 6, BusinessWeek Online Reporter David Shook spoke to Barker about the progress his company has made since the government-funded Human Genome Project delivered its first draft of the complete genome sequence in 2000 -- and how much his industry still has to accomplish. Here are edited excerpts of their conversation:

Q: How far has the biotech industry come in developing genomics-based drugs since the human genome was sequenced a few years ago?
A: There hasn't been a whole lot of impact yet. No drugs have been approved by the U.S. Food & Drug Administration as a result of the genome efforts. Pharmaceutical companies now have many more drug targets than they know what to do with. For many years, the pharmaceutical industry focused on about 500 drug targets. All of a sudden, we have several thousand more available because of the high-throughput (HTP) screening of genes that companies such as ours have been doing.

Q: By drug targets, do you mean genes that play a role in disease?
A: I mean the proteins that play a role in disease and can be influenced with a drug. Genes code for the creation of their own proteins. And we're trying to understand the function of all these proteins that have been discovered through the process of analyzing gene expression. Some proteins do bad things.

If you identify an active gene that seems to play a role in disease, the next step is identifying the protein that the gene makes. From there, you may be able to find a small-molecule drug that binds to the protein and inactivates it if it's doing something harmful.

Q: Is it as simple as finding a drug that turns on or off the activity of a certain gene?
A: It's usually not as simple as one gene being turned on or expressed abnormally that leads to a disease...[like in] sickle-cell anemia, where one slight genetic variation causes the disease. It's usually 5 or 10 gene variations working together. And if you find a potential drug for a disease, it may have an effect on entire [cellular] pathways in the body -- consequences you may not see at first.

Genotyping [the study of gene variations or mutations] will help unravel that puzzle. That's where technology such as ours can be useful. It just will take time.

Q: How exactly is your company shedding light on the activity of genes and their proteins?
A: The main tool we're developing is for genotyping -- determining the variation in people's DNA, as well as how that variation causes drugs to be effective, not effective, or even toxic to people. It's a nasty little secret that adverse drug reactions cause 100,000 deaths a year in the U.S. and put 2 million people into the hospital annually. Pharmaceutical companies have to take into account genetic variation among people to understand how drugs can be helpful, harmful, or have no effect.

Prozac, for example, is ineffective in about a third of people who take it for depression. Nobody is sure why. If the doctor knew that a particular patient would get no benefit from Prozac in advance of treatment, that would save lot of money for everyone. But so far, the pharmaceutical companies have been reluctant to do large-scale genotyping for the patients in their drug clinical trials. Many companies still consider it too expensive.

We're bringing down the costs associated with genotyping -- from an industry standard a few years ago of about $1 per test to a few cents per test pretty soon. Pharmaceutical companies know they need to do genotyping on a larger scale, but many companies haven't fully committed to it yet.

Q: Who has committed to this type of research?
A: GlaxoSmithkline (GSK ), the largest European drugmaker; Merck (MRK ); and Bristol-Myers Squibb (BMY ) have shown more commitment than most.

Q: Right now, genomics companies aren't profitable and many of the genomics stocks are at or below $5 a share, including Illumina's. Does that hurt your ability to raise money? Do you think investors will jump back into these stocks eventually?
A: First of all, we don't need to raise capital. We were fortunate to raise money before the crash in this sector. [Illumina has $80 million in cash, according to its latest balance sheet.] Second, I really do think there is value in this industry for investors. I think we'll see tremendous benefits from genomics for the biotech industry in the near future. But right now, just about every stock in the industry is down because there haven't been any drugs created yet from all this new technology.