Which cliche do you prefer-the "book of life"? The "blueprint for humankind"? How about "the code of codes"?
Unless you just woke up from a week-long sleep, you've been hearing all of the above, and more, in the wake of Monday's announcement that the Human Genome Project, funded by the U.S. government and a British charitable foundation, and a parallel effort by the private biotech firm Celera Genomics, had finished the rough-draft map of the human genome. Kudos all around: it's a singular achievement.
But excuse me if I keep my champagne on ice for a little while longer. What has actually been completed here is less than what most commentators are hyping. And even when scientists do achieve what people think they have already, the brave new age of medicine and health will not be just around the corner.
So what have the public and private efforts actually accomplished? At the risk of coining my own cliche, I prefer to think of it as peeking at the source code for Homo sapiens. You know-something like the 0's and 1's that make up the machine language of every computer. But instead of 000101010001100--on and on, what the scientists have done is produce a readout like this: AATCGTTAACT--for, oh, 3.1 billion or so such letters.
Each letter represents one of the four molecular units (dubbed A, T, C and G) that make up DNA. My point is that this string is even less understandable than those 0's and 1's.
Although a handful of computer jocks might be able to glance at the binary code and pick out, say, the instructions that say "add 1 to the previous answer," that's about all genome scientists can do at this point. They have little idea what those strings of letters mean. They don't even know how many genes they constitute-guesses range from 40,000 to 100,000 and up.
Not to belabor the analogy,but they don't know how many different instructions the billion-number-long sequence of 0's and 1's breaks down to. All of which is to say there's a lot of work-decades of work-to be done before we know what this babel of A's, T's, C's and G's actually means.
Of course, we don't have to know the meaning of the entire genome before we can exploit some of it for human health. But that's been happening all along: scientists have figured out which genes are active in one form of leukemia but not another, for instance, allowing them to make a more accurate diagnosis than has been possible with traditional pathology. An accurate diagnosis helps when you're trying to figure out what chemotherapy to give the patient.
We have tests for familial colon cancer, inherited breast cancer, Huntington's disease, sickle cell anemia and scores of other genetic diseases. All of this happened without knowing the genome. And, sure, it's wonderful--except for one thing: carrying the gene "for" some dread disease doesn't necessarily mean you'll contract the disease (unless it is a so-called single gene defect disease, like sickle cell or Huntington's but unlike cancer, heart disease, diabetes or mental illnesses).
At least 20 percent, and maybe more, of the women who carry "the" breast cancer gene (BRCA1) do not get that disease. Half of all identical twins, who have identical genomes, do not get schizophrenia even though their twin does. Clearly, something besides genetics is at work here. It would therefore be a mistake to think that once we have deciphered all those A's, T's, C's and G's, and turned it into a test we give newborns, we will know exactly what diseases that child will contract.
If diagnosis is iffy, what about treatment? We have known for years exactly what gene causes Huntington's disease and what gene causes sickle cell. Again, these are pretty simple, single-gene illnesses. But we haven't come up with a cure, let alone a preventive, for either. Do we really think cancer, or heart disease, is going to be easier?
Call me a cynic, but I'm also skeptical that the first large-scale use of the new genomic knowledge will be in diagnosing or treating dread diseases anyway. My money is on designer babies. Once enough of the genome has been deciphered so that we know, say, that this 35,000-letter-long sequence spells out the gene for a particular form of a dopamine receptor that supposedly makes you a risk-taker, I can see parents choosing their kid's traits like options on their SUV. Tall: check. Thin: check. Shy: forget it. Good memory: check.
After all, we are the society where you can make a living selling sperm from Nobel laureates, and other brilliant men, to couples seeking donors for in vitro fertilization. Let's say prospective parents start telling the in vitro clinic to slip in the gene for the receptor that improves memory, and the gene for the neurotransmitter associated with being a risk-taking dot-com entrepreneur.
My question is this: what happens to the kids who don't turn out to be what mom and dad ordered? For that's what will happen: if having the breast cancer gene does not necessarily mean you'll get breast cancer, what makes us think that the gene associated with even more complex and environmentally-influenced traits like personality and intelligence will turn out to be deterministic? Are we going to let parents send the kid back, and get a refund?
The curtain on the age of the genome is rising. None of us can say what will now happen on the stage. But I'll wager that it won't be the golden age of medicine that this week's hype has been promising.