Dr. Paul Billings, MD, PhD, is Vice Chair of the Board of Directors of the Council for Responsible Genetics and Chief Medical Officer of Life Technologies, Corp. The following is excerpted from an interview and represents Dr. Billings' own views rather than those of Life Technologies.
To the extent that one's genomic DNA is stable-and I believe that the vast proportion of anyone's DNA at any particular time is in fact stable, and does in fact basically reflect what you inherited at the time of conception-the analysis of that over the next 20 years will become increasingly simple and very inexpensive. And to the extent that this information is a reliable, quantitative, and identifiable component of disease diagnosis, therapeutic selection, et cetera, I believe that information will become an integral part of everybody's medical record, probably from conception and certainly from birth.
Genomic DNA information has very distinct advantages: it is measurable, I believe it will be highly reliable, and it is for the most part stable. Epigenomics, post-translational variability and environmental influences can be important and do modify the genomic information. And of course disease states like cancer are characterized by finding more genomic mutations than in non-cancerous cell genomes. But the overall impact of the genome should be reliable and accessible. We'll figure out how to use that over the next 20 years, along with environmental impacts and other kinds of more variable components of our biology, to make more accurate, more reliable diagnoses, and to make more biologically formed choices about treatments and prevention.
The fundamental thing that will be different in 20 years is that our medical records will be built more significantly on genetic and genomic information, and will be verifiable in a way that our current medical systems are not. That is not to say that genetics or genomics is the be-all and end-all of risk-it's clearly not. Genetic risk is highly environmentally modifiable, and even though the genome is for the most part stable, mutation does occur and modification of the expression of mutation can be significant. The genomes of cancer cells are somatically mutated at an amazing rate, and clearly there are epigenomic effects and modifications that can influence the power of a particular germline-inherited pattern.
You still have a certain set of genes at conception, and we can elucidate very accurately and cheaply what those genes are, along with many other factors that make genome analysis more complex and add a more nuanced and personalized story. Much of that is cultural and environmental, but there's still a big difference between asking questions about your family medical history and sequencing your genome to find out whether you have this gene or that gene. That will be a major and on the whole useful change.
Today there are many kids who are born with syndromes, but we don't have any idea what's wrong. We know that there's something wrong, but we really don't know what the bases of many birth syndromes are. We're going to find a lot of these in the genome. Not all of them, and some will be methylation or exposures to mutagens in utero, but we're going to have a more concrete basis for building up that knowledge than we used to. There are a lot of kids who die in the first year of life from disorders which we'll identify either in utero or at birth, and we will prevent many of those deaths.
The $100 (and $10) genome:
I see in the next 20 years the technology driving down the cost of genomic sequencing to being very inexpensive, and it's dramatically going to continue to drop. The $100 genome will be available in 5 years, and the $10 genome in 20. And at $10 per person, almost everybody in the world can get it. That will be the cost of generating the raw sequence; interpreting it and making it useful to the individual will add costs, maybe substantial ones.
Highly medicalized analysis will take more time and will be more expensive, but a gross, computerized look at your genome will be available for very little money. As we do more DNA sequencing and we have more ability to correlate people's sequences or clinical information-as medical records get better and people begin to share more information, and get involved in registries and research projects-the ability to automate analysis goes up. This means that over 5 or 10 or 15 years, the automated analysis-I would call it the "gross analysis" of one's genome-is going to get much better. You'll still want to go to an expert, you may want multiple looks, and there will be lots of other things that will help modify the interpretation beyond a single computer algorithm; but the trend will be toward automated analysis, and with that the cost will go down.
The limits of genomics:
Now, there are a lot of things that the genome won't be that important for. I remember when there was talk about the genetics of leprosy. There is an element of genetics to leprosy, as it turns out, but it's essentially an infectious disease. There may be genetics about what kind of manifestation you have or how severe a case you get, but the bottom line is that you have to have "the bug around" to get leprosy. There are going to be a lot of diseases where the major risk factor will always be, do you live in an environment that harbors the infectious agent? In some conditions, the kind of microbes that reside in your gut, and the proteins they produce, may interact with your body (and there may be some genetics to this) to cause illness or modify severity of conditions like autoimmune disorders.
In many cases, your gene combinations might only confer moderate risk for a disease, and it's the environment around you that matters more. But even if it's just a small percent of the population that has powerful genetic determinants, to help that small percent is really a major deal.