A decade ago, everyone thought that sequencing the first genome would lead to an astounding number of cures for diseases. Human Genome Sciences'
Three years ago, it cost $1 million to sequence a human genome. Stock Advisor pick Illumina's
In addition to producing cheap sequences -- well, "cheaper," since it's still out of my price range -- the HiSeq 2000 is also fast. The machine can produce data about 3.5 times faster than Illumina's latest update to its older Genome Analyzer machine.
With competition from upstarts like Pacific Biosciences and larger companies like Roche and Life Technologies, Illumina needs to get machines into the hands of researchers as quickly as possible. In addition to selling the sequencers, Illumina sells the reagents to run the machine -- a nice razor-and-blade model. At a cost of more than half a million dollars for each machine, researchers are likely to continue using the HiSeq 2000 even after another company develops a faster and cheaper model -- but only if they've actually bought a machine. Illumina is off to a good start; 128 machines are already sold to an international genomics institute, BGI. They're expected to begin shipping next month.
The real winners
While yesterday's announcement is a great one for Illumina -- until another company comes along and beats it on speed or cost -- the real winners in the quest for cheap sequencing are the people who will use the sequences.
The initial beneficiaries will be companies like Knome, Navigenics, and 23andMe, which analyze customers' genomic sequences to tell them what their DNA says about their characteristics. Thousands of genetic differences have been linked to certain traits -- medical ones like susceptibility to cancer, and more benign ones like the type of earwax you have -- but sequencing additional genomes is bound to help researchers discover new genetically-linked traits.
As more medically relevant traits are discovered, the fields of electronic medical records and sequence analysis will collide. Expect Google
Down the line, drug companies will be the true beneficiaries of cheap sequencing. If you look at the profile of many drugs, you'll find that they work well on a subset of patients and don't work well for other patients. Some of that may be due to chance, but there's likely a genetic component to it as well.
For instance, a variation in a gene doesn't convert Bristol-Myers Squibb's
In both those cases, the genetic differences that cause decreased effectiveness were discovered after the drugs were on the market. Essentially, the drugs worked so well that they were able to show a clinical response even when nonresponsive patients reduced the average effectiveness.
But as more drugs and better drugs are developed, that average response rate continues to increase, and drug companies need to clear an ever-higher standard. Removing non-responders from the trial can help them do just that. Today, discovering genetic variations that result in non-responders can be costly, but sequencing the genomes of a thousand patients in a clinical trial won't be that expensive just a few years from now.
Keep an eye on the cost of sequencing, Fools; it has the potential to make you rich faster than you'd expect.
Editor's note: A previous version of this article misstated the HiSeq 2000's price. The Fool regrets the error.