Over the last ten years, a great shift in research prerogatives has taken place. Armed with knowledge of the roughly 22,000 human genes and other interspersed genetic elements that make up the human genome, scientists have become focused on identifying how this genetic material influences human disease. To this end, research has become preoccupied with genetic variants—that is, individual variations in the genetic code—for the roles they may play in disease, and a body of compelling data has been collected in this area. The use of methodologies like genome-wide association (GWA) studies has certainly led to progress in defining the possible genetic basis of a variety of conditions. But with this progress has come the realization that uncovering common genetic variants may only be scratching the surface.
Hundreds of GWA studies have been published over the last ten years examining the contributions of common variants to a variety of diseases, but according to David Goldstein, most of the work in understanding genomic control is yet to be done. Goldstein—who is director of the Center for Human Genome Variation at Duke University and a molecular genetics and microbiology professor—is now looking beyond the common genetic variants, and instead focusing on what is rare. He is not the only one, either. “Although there are differences in view about just how successful a GWA study is or isn't,” he says, “it is now clear and accepted by all that, at best, it has revealed only a minority of the genomic control over disease.”
Shared by fewer than 3% of people in the population, individual rare variants, or combinations thereof, are a promising new direction and may answer the genomic control question that researchers like Goldstein are searching for. But their low frequency in the human population makes identifying these rare variations a challenge.
“The problems that still need to be overcome for finding rare variants have to do with resolution of microarrays and their inability to find all of the rare variants, either directly or indirectly,” says Stephen Scherer, director of the Center for Applied Genomics at Toronto's Hospital for Sick Children and professor of molecular genetics at the University of Toronto. Microarrays—which have been the key tool for conducting GWA studies up to now—can only detect variation that has been designed onto the array. Therefore, rare single nucleotide polymorphisms (SNPs) or copy number variations that might be involved in a particular disease would be missed.Tools of the Trade
“GWA studies have been the primary workhorse in the search for variation, but at its heart, GWA is designed specifically for common variation,” says Goldstein. “The whole paradigm works by identifying the polymorphisms in the human genome, and until recently it was the only technique that was comprehensive for variation.”
The microarray platforms used in GWA look at gene expression or structural changes within the genome. Tiling arrays, which are microarrays composed of short probes that can span an entire genome, can evaluate known and unknown genes, and are manufactured by companies including Affymetrix, NimbleGen, and Agilent. On average, those arrays based on photolithographic manufacturing can hold up to six million discrete features and each contain millions of copies of a single probe. Tiling arrays based on mechanical spotting or printing can hold an average of 400,000 features.
To accurately analyze any variant, researchers need large sample sizes to obtain statistical significance; to this effect, microarrays are designed specifically to analyze the target variant in high numbers of people at a low cost. But because microarrays must be programmed in advance to search for a specific known variant, these arrays are not a discovery tool. This has hampered their usefulness when it comes to rare variant identification.
The Next Generation of Variants
Now, when it comes to finding diseaseassociated rare variants, many scientists are pegging their hopes on a technology that has exploded in recent years: next-generation sequencing. The technology provides a means to analyze greater numbers of whole genomes in order to search for rare variants, but it is bogged down with voluminous data and technical limitations. Furthermore, sample sizes—though expanding—are still restrictive. In the face of these drawbacks, researchers are currently trying to determine how to better GWA studies by integrating the advantages of next-generation sequencing with microarrays.