28 May 2012
It’s not every day that your mind is blown. I would like to share with you an epiphany that happened to me not too long ago, dear reader.
Last week I had the pleasure of hosting my graduate student symposium. In my program, every student is required to give a 15 minute talk encapsulating their research into a cogent, concise message. Along with our student talks, we had two invited speakers. Jonathan Sebat is a professor of Psychiatry at UCSD and has garnered glimpses into modern-day human evolution that were unthinkable just a few decades ago. The method? Nothing less audacious than whole-genome sequencing.
For a moment, let’s pause and think about that. A genome is the sum of the sequences of all of the DNA of an individual. The amount of information is so large that it is faster and easier to send genomes via snail mail instead of the internet. If one were to recite one base of DNA (A, T, C, or G) per second of ONLY the 5% of DNA that encodes proteins, it would take someone about 4 months of night and day continuous singing to sing their whole biochemical song.
So, how long does it take to sequence a human genome these days? A day or two. How much does it cost? About $1000. The magnitude of information and the scope of this technology is admittedly difficult for this young scientist to wrap his head around.
In a previous post, I talked about the rise of individual cancer genomics and how sequencing the DNA of a cancerous tumor could literally find the needle in the molecular haystack that could reveal a deadly cancer’s achilles heel.
Unfortunately, such technologies have revealed a somewhat dubious prediction about cancer: cancer cells in a tumor are very heterogeneous. In an article in the New England Journal of Medicine, Gerlinger et al sequenced multiple regions of the same tumor! What they discovered is that molecular diagnosis (and cancer treatment) can vary by the region of the tumor that a doctor pricks for a biopsy. Even worse, it predicts that many cancers would be resistant to new-age selectively targeted therapies. Not good.
During the symposium I had the pleasure of having lunch with Dr. Sebat of UCSD and Dr. Wolf of Sanford-Burnham. I expressed my dismay about the aforementioned, and Dr. Sebat pointed out something that I had not considered: not too far in the future, it’s likely to be cost effective to sequence many regions of a patient’s tumor and identify the mutations that actually contribute to the disease, as opposed to “passenger” or “private” mutations that only arise because the tumor is genetically unstable and are not actually contributing to the disease.
In other words, we will have the technology to turn cancer’s incredible resiliency against itself. Doing so, we will be able to use (and further develop) selectively targeted therapies that will enable cancer to be treated much more effectively with less side-effects, greatly improving the human condition.