Happy Birthday, human genome. On June 26, 2000 a group of scientists at the White House announced that they had a working draft of our genetic blueprints. They hadn’t sequenced all our genes; the Human Genome Project and its private-sector competitor Celera Genomics still had some gaps to fill in. Still, scientists believed this data might hold clues to the causes of certain diseases and could lead to new treatments.
Even before the project’s start, some scientists were skeptical: Was mapping our genome a waste of money and time? Even among public hoopla and presidential speeches, scientists cautioned that applying the results would take time. Now, ten years later, many are asking: What have we learned? Here we round up some opinions about the impact of the project.
The Bad?
Some see fewer medical treatments than advertised. Instead of simple relationships between common variants and specific diseases, sequencing uncovered sheer complexity. Researchers now think that intricate relationships between rare variants may cause many diseases.
The difficulties were made clear in articles by Nicholas Wade and Andrew Pollack in The Times this month. One recent study found that some 100 genetic variants that had been statistically linked to heart disease had no value in predicting who would get the disease among 19,000 women who had been followed for 12 years. The old-fashioned method of taking a family history was a better guide. Meanwhile, the drug industry has yet to find the cornucopia of new drugs once predicted and is bogged down in a surfeit of information about potential targets for their medicines. [The New York Times]
As genetic sequencing goes, what once took years and millions of dollars can now take months and thousands. Still, some worry that the drive to sequence more, faster has led to techniques that make reading results increasingly hard.
The advances in speed … have come at a cost. Only short stretches of DNA can be sequenced at a time, so the pieces have to be joined together by looking for overlaps between them. While early instruments sequenced pieces up to 900 base pairs long, most high-speed machines produce “reads” of less than 100 base pairs. That means the overlaps are much shorter, making it far harder to join the pieces together, so assemblers use existing genomes as a guide — which can lead to mistakes. [New Scientist]
The Good?
Though the Human Genome Project may have thus far yielded fewer advanced medical treatments than hoped for, the findings for biologists seem greater than expected. The complexity that frustrated those looking for practical, clinical applications has led to rich veins of research.
Nature News surveyed more than 1,000 life scientists, many who said that the sequenced genome had greatly benefited their work:
Almost all biologists surveyed have been influenced in some way by the availability of the human genome sequence. A whopping 69% of those who responded to Nature ‘s poll say that the human genome projects inspired them either to become a scientist or to change the direction of their research. Some 90% say that their own research has benefited from the sequencing of human genomes — with 46% saying that it has done so “significantly”. And almost one-third use the sequence “almost daily” in their research. “For young researchers like me it’s hard to imagine how biologists managed without it,” wrote one scientist. [Nature News]
Some also praise the accessibility of genomic data from this research as a means to advance further research–among them, Francis Collins, current NIH director and former head of the Human Genome Project.
“For example, the search for the cystic fibrosis gene finally succeeded in 1989 after years of effort by my lab and several others, at an estimated cost of U.S. $50 million,” Collins writes in an opinion piece published in this week’s issue of the journal Nature. “Such a project could now be accomplished in a few days by a good graduate student. … ,” he writes. All the budding geneticist needs, Collins says, is the Internet, some inexpensive chemicals, a thermal cycling machine to amplify specific DNA segments, and access to a DNA sequencer, which “reads” DNA via light signals. [National Geographic]
Nature News‘ poll also hints that scientists believe that a better understanding of the underpinnings of human genetics, better systems to analyze the sequenced data, and more information from other research like the Human Epigenome Project will help turn this biological knowledge into clinical applications–some argue within the next 10 to 15 years.
Others say we can only wait and see. That’s what Eric S. Lander, director of the Broad Institute, told The New York Times regarding a direct connection between sequencing and treatments.
“The only intellectually honest answer is that there’s no way to know,” Dr. Lander said. “One can prefer to be an optimist or a pessimist, but the best approach is to be an empiricist.” [The New York Times]
Related content:
80beats: Court Strikes Down Patents on Two Human Genes; Biotech Industry Trembles
80beats: IBM’s “DNA Transistor” Could Sequence Genomes on the Cheap
80beats: New Lawsuit Challenges the Patenting of Human Genes
80beats: Big Autism Study Reveals New Genetic Clues, but Also Baffling Complexity
Image: flickr / net efekt
June 25th, 2010 at 8:45 pm
We can never have enough data. We may not know what to do with it yet, but it is all valuable. Even data with no apparent value still has value in knowing that it’s valueless, and should be kept as example so the territory isn’t re-covered.
But knowing how computers work intimately separates the professional programmers from the amateurs.
And how long was physics around before Einstein’s relativity and theories shook the world? There *must* be value in DNA study, because it literally dictates everything about how we should be built and operate, hardware wise. It’s like saying the blueprints to a building don’t have value because you can’t read them properly.
Only time will tell, this is true, but time has told again and again, the more information we have the better we are able to cope with things – we just can’t predict in what ways beforehand.
It took hundreds of millions of years to write this code, mind you, and 50 years to unlock since the discovery of DNA, we shouldn’t be surprised at all that great things haven’t happened yet. Come back to me when the human genome is 50. :)
June 27th, 2010 at 6:43 pm
Let’s face facts. The diseases with simple genetic causes, I believe these are referred to as SNP’s, were already mostly understood long before the detailed genetic data came in. Sure, medicine may not have known what the exact “normal” and exact “abnormal” sequences were, but doing the Mendeleevian inheritence maps zero in on these fairly quickly. Once you have appropriate expertise and the right data set available.
Mostly, what is left are the complex diseases. Ones with multiple genetic loci, or ones with intertwined genetic and environmental causes.
If anything I am more confident and optimistic than @1 nick is. The real question will be, once we start gaining deep understanding of our DNA, will we use the knowledge wisely?
June 29th, 2010 at 1:21 pm
Having a complete list of all the proteins possible to be made in the human body seems pretty invaluable. I think the big thing holding us back is that exact protein structure is really hard to determine from the DNA sequence. Things have been getting better, and I have heard of things like high throughput x-ray crystallography are on the horizon, but without it, its like having the box of parts for a car without a picture of the finished product. Predictions can and are being made, but since drug research typically requires exact structures to look for potential effector molecules, the genome sequences aren’t the magic bullet for figuring out stuff in the computer just yet. There is a lot more to be learned before the genome can be transformed into a living organism on a computer, but its a big first step, and its now public and free.
That being said, I’m pretty sure all medical and biological sciences benefit from the HGP. I use NCBI’s alignment programs all the time, for free. How can you identify mutants if you don’t know what the original should look like? In many ways, I think the HGP was more valuable than the moon mission or other such prestige undertakings, since it informs every facet of biological sciences.