Complete Genomics, by Radoje Drmanac
June 1, 2011
Complete Human Genome Sequencing — Where Is It Headed?
Flashback to 1989: Rade’s vision was to sequence a complete human genome on one array.
Updated to clarify point pertaining to clopidogrel
As chief scientific officer at Complete Genomics, I spend the majority of my time advancing complete human genome sequencing technology and its applications. Genome sequencing has been a passion of mine for almost 25 years, ever since I used my first hybridization-based DNA sequencing assay back in Belgrade, Serbia, in 1987. Even then, I wanted to be able to sequence a complete human genome on a single DNA microarray using microbeads.
At Complete Genomics, we decided to focus on sequencing complete genomes from day one. Partial genomes provide incomplete data that is insufficient for the successful interpretation of most genetic variants, as their meaning is contextually dependent on other variants in the genome. Current exome sequencing technologies, for example, not only miss a significant fraction of exons but also miss all the noncoding variants and structural variations. Also, because of the significant sequencing and computing complexity associated with complete genome sequencing, we realized from day one that as this market develops it will largely be addressed via outsourced service models.
Our current customers range from academic centers and government institutions to biopharmaceutical companies. We enable them to conduct large studies of complete human genomes quickly and cost-effectively.
We are continuing to scale up our commercial sequencing operation. I am particularly proud of the efficiency of our imaging system for our patterned DNB™ nanoarrays that currently use only two pixels per DNA spot and will shortly start using 1 pixel per DNA spot. This is a maximal possible imaging efficiency. Our imaging instrument currently generates one genome per day, and next year will generate several genomes per day. With such efficiency it becomes feasible to sequence millions of human genomes per year for clinical use.
Our sequencing service has already been used to advance medical research in therapeutic areas as diverse as oncology, Alzheimer’s disease and hypercholesterolemia. Despite the extreme complexity of regulatory networks, I expect our understanding of the genetic basis of disease to grow quickly. The number of sequenced genomes is increasing exponentially, and our genomic knowledge grows exponentially with every sequenced genome. A few patients with rare diseases have already been helped, which is just the beginning of a far larger change in medical practice. In the future, patients with diseases that are difficult to diagnose will have their genomes sequenced immediately to help their physicians understand their condition. Furthermore, complete genome sequencing will dramatically improve diagnosis of genetic diseases during routine IVF procedures.
I anticipate that sequencing genomes from cancer biopsies will revolutionize cancer treatment for the several million new patients diagnosed each year. If I am diagnosed with cancer, I will not let my doctor treat it without sequencing my tumor genome first. In the future, I can also see us sequencing cancer genomes as a preventive screening measure. Recent research has shown that a cancer usually takes a decade to develop and starts shedding cells before it produces any symptoms. People could simply get a blood test every two or three years to check for circulating cancer cells. If cancer cells were detected and sequenced, the clinician would have knowledge that would help stop the disease before it develops further.
In the coming years, I foresee routine complete personal genome sequencing as the “universal genetic test”. It is a noninvasive, comprehensive genetic analysis that subsumes all disease-specific genetic tests and produces data that a person can mine throughout their lifetime to improve health. Although clinicians may not be able to interpret all this data now, and we should be careful to avoid overinterpretation, this data will be ready and available when needed, and as our understanding of the genetic basis of disease continually develops in the future. For example, there are genetic variants in the CYP2C19 gene that determine when you are having a heart attack whether you should receive the standard drug clopidogrel or need to receive a different one. That important decision needs to be made within 30 to 40 minutes, which is challenging even for targeted diagnostics. But if you have already had your genome sequenced this information could be available in less than a minute.
This genomic medicine revolution will require lots of technology and software development, additional basic and applied research, and changes to both medical practice and education. I am excited to see it happening.
posted on July 11, 2011