With continuous improvements in read length—10 kb and beyond now a reality—higher throughputs, and declining costs, Dr. Nickerson concluded that single-molecule sequencing will propel the field closer to the goal of the $1,000 genome. She added, however, that this figure accounts only for the cost of generating the sequence and not for the expenses associated with storage and data analysis. Dr. Nickerson described recent studies in which WGS is being used to identify somatic variants associated with cancer and exome resequencing projects aimed at identifying gene variants in protein-coding regions associated with Mendelian disease.
Euan Ashley, DPhil., an assistant professor of medicine and director of the Center for Inherited Cardiovascular Disease at Stanford University, talked about how dramatic declines in the cost of genome sequencing will increase the data generated exponentially, as “thousands, if not tens of thousands” of genomes are sequenced in the next couple of years.
He described the recent work of a Stanford team focused on extracting clinical value from genomic data (Lancet 2010;375:1525-1535) linking rare variants to disease risk, and applying the results of GWA studies to individual patient genomes.
Dr. Ashley emphasized the need to reconfigure the current databases. Most of the genomic databases are basically catalogs of genes and variations. They need to be reformatted in a way that would make them easier to interpret and use in a clinical setting, he urged. Furthermore, he pointed to gaps in the existing information; in particular, the regulatory and non-coding genome “has been neglected,” he said. Another challenge in the field at present is learning how to explain genomic data to patients.
Russ Altman, M.D., Ph.D., chair of the department of genetics and bioengineering at Stanford, discussed the utility of the Pharmacogenomics Knowledge Base PharmGKB, describing it as being “like Amazon for drugs and genes.” It allows users to search for genes and variants associated with drug dosing and efficacy, for specific drugs and related genomic data, for diseases, and for pharmacokinetic and pharmacodynamic pathways, all with supporting literature.
In his presentation Dr. Altman described the use of PharmGKB to link a heterozygous null mutation in the CYP2C19 gene (a member of the cytochrome P450 system of enzymes that determine drug metabolism) identified in an individual’s genome to a 50% reduced capacity to metabolize certain drug types. This information would be used to drive drug selection for this particular individual.
Moving from the level of individuals to a broader view of how to use genomic data to reduce the burden of disease, close the gap between discovery and application of genomic information, and track outcomes, Cecelia Bellcross, Ph.D., a fellow with the Office of Public Health Genomics of the Centers for Disease Control and Prevention, noted the paucity of funding to support the back-end of the genomic medicine pathway—comprised of clinical testing, data interpretation/application, and outcomes—compared to the funds available for generating genomic data.
Noting that every individual genome is likely to have “abnormalities”—variants that may or may not have clinical significance—Dr. Bellcross underscored the need to tackle this “evidence dilemma” in genomic medicine in a systematic way and not on an individual basis. Strength of evidence does not imply strength of association, she cautioned, and genetic association does not necessarily equate to clinical risk.
Dr. Bellcross urged stakeholders—including funders, patients, clinicians, academicians/researzchers, and lawyers—to work together to close the gap between evidence and clinical applicability. The public-private partnership known as GAPPNet (Genomic Applications in Practice and Prevention Network; is an example of a strategy to streamline the use of validated genomic knowledge.