The new millennium brought us the culmination of the Human Genome Project, but this didn’t mean that there was one single human whose genome was decoded. Rather, it was a deliberate combination of adults, most of whom were unidentified volunteers from Buffalo, New York, along with J. Craig Venter, who jumped in and churned up the water, but that’s another story.
It’s taken until 2008 for any individual to have his DNA completely decoded and publicly published. The first four are: James Watson of Watson & Crick fame (sans two biomarkers which are currently believed predictive of Alzheimer’s), J. Craig Venter again (who most agree probably beat Watson to the digital punch but held back for history’s sake), an unidentified Asian male, and an unidentified African male.
One might ask why this is important. And it’s a good question. The truth is: It’s not. Not yet, any way.
Think of it as the biotech equivalent of Metcalfe’s Law. Bob Metcalfe is a well-known computer scientist, who co-invented the Ethernet and founded the communications company 3Com. He’s been all about networks for decades and he formulated Metcalfe’s Law as follows: “The value of a telecommunications network is proportional to the square of the number of connected users of the system.” For most of us, it means this: The value of one telephone — where you can call no one — is zero. The minute someone else gets a telephone, then the two of you can talk and the value of all telephones go up. As more and more people get telephones, the value of the whole network goes up. It’s an idea anyone can understand.
Getting back to this whole-genome decoding of human DNA, it’s not the same. The decoding of the second one only marginally increases the value of the first, if it does at all. We see the DNA decoded all right, we see the raw data… but we don’t know what it means yet. In the meantime, we’ve decoded smaller swaths of human DNA, and we’ve discovered quite a lot. We once thought we got one set from Mom and one set from Dad. It turns out that we sometimes get multiple copies of strings of DNA from the same parent. Could these variants be the “nature” part of the “nature vs. nurture” equation… the reason why each child of the same parents seems to be a lot more different than alike?
A team of Swedish scientists from Karolinska Institute discerned that men with the “334 version of the AVPR1A gene” had trouble committing. They were less likely to be married, and if married, were more likely to report having marital problems. Having two copies of this DNA only made things worse in the marriage and serenity department.
And thus, a little bit of DNA information leaves open the door to a lot more inquiry.
But when we talk about “whole genome” DNA … and a whole planet … how much data would that be? The calculation appears to be simple: Each human has some 3 billion base pairs of DNA, each represented by a letter. That would be 6 billion letters, or 6 gigabits of data. Your own DNA data would fit on thumb drives already floating around in your possession on the memory chips of your digital camera, but that adds up when we get to all the humans on the planet. In round numbers, there are 6+ billion folks, and that makes 6 billion humans times 6 gigabits, and that’s a whole lotta data.
A team of Swedish scientists from Karolinska institute discerned that men with the “334 version of the Avpr1A gene” had trouble committing.
Then we have the big presumption that we have the technology to store all that data, and to analyze it, and that in doing so, there would be value.
Which returns us to the question: Does it follow Metcalfe’s Law? Does the value of the human network of decoded DNA grow with every human fully decoded?
It would seem not – the value of the second telephone is only compromised if we don’t want to talk to the second person, but we are quickly motivated to find someone we want to speak with online. With DNA, it doesn’t appear to work that way. In fact, that’s possibly been proven by whole-genome decoded Human #5.
This time it’s a woman, and it’s completely different. Read the November issue of Nature if you want the details, but the basic story is this: A woman in her fifties had developed acute myelogenous leukemia (AML), a very aggressive cancer. Washington University scientists decoded the DNA of her tumor cells and then also decoded the DNA in her normal cells. And why? The clue here is that whenever a cell divides, it can mutate. And that mutation can be carried forward every time the cell divides from then on. Over a lifetime, particular cells in your body can mutate a lot. When you have cancer, there has been a sequential series of mutations in some of your cells, which, when combined with your starter DNA, has created a formula in which your cells can — in the vernacular — run amuck. That’s right. If you have cancer, the DNA in your tumor cells are different from the DNA in the rest of your body. So, we humans are not just a single set of DNA. Mutations happen. Pair them up with your starter DNA, and the result can be benign… or deadly.
That’s why the decoding of Human #5 was a lot more interesting — for the first time, scientists examined two complete sets of DNA within the same person.
Previously, two gene mutations were associated with AML. After decoding this woman’s DNA and performing a whole lot of sophisticated computer analysis, they have uncovered 10 mutations. Three are in genes known to suppress tumor growth — that’s not good. Another four mutations were found in genes known to promote cell growth — which is bad if it’s a tumor cell. Another gene basically strong-armed the chemotherapy generally prescribed for these patients, so the patient would suffer all the side effects of the treatment and little, if any, of the benefits. Why this set of DNA mutations created a dire situation became eminently clear.
To see if they could generalize, the scientists looked at 187 other AML patients — none had these new mutations. They suspect that cancers may be very specific to the individual. They believe that cancer cells develop successively, mutation by mutation, until some particular moment when a cruel combination of new mutations and starter DNA arrive together and a catastrophe of cell growth occurs.
So what does this mean… at least on a digital data level? Without a doubt, we are now experiencing the Big Bang of Biotech Data. Not because we can simply create all this DNA data, but because we need to create all this data to figure out who we humans are and how we tick. But unfortunately, we don’t know what data we need for what, and what — in the end — will prove useful. We are still shooting in the dark.
So, while with the simple telephone example, we can easily have a sense that Metcalfe’s Law is true for networks, it doesn’t quite translate to DNA.
But that doesn’t mean we can’t have a law about it. I call it “Moira’s Law”:
“The value of decoding human DNA is proportional to the number of people who will benefit.”
I’m sure I’ll be changing it as the bio age unfolds, but in the meantime, it seems to me like a good way of thinking about it.
Moira A. Gunn, Ph.D. hosts “BioTech Nation” on NPR Talk and NPR Live. She’s the author of Welcome to BioTech Nation … My Unexpected Odyssey into the Land of Small Molecules, Lean Genes, and Big Ideas cited by the Library Journal as “Best Science Books of 2007.”
Copyright 2008 Moira A. Gunn