Technique Uses Linkage Analysis to Detect Gene-Gene Interaction in Disease
(From the May 2014 issue of Research Now)
Scientists have developed a linkage-analysis-based technique that detects interactions between multiple genes involved in disease with greater accuracy and detail than conventional methods. The research could aid in the discovery and understanding of genetic causes of complex diseases such as epilepsy. The work, co-authored by David A. Greenberg, PhD
, in The Research Institute at Nationwide Children’s Hospital, was published online in April in PLoS One.
Over the past decade, more and more scientists have turned to a technique called Genome Wide Association Studies, or GWAS, to identify specific genes involved in disease. While this method was expected to identify important genes for common disease, its results have been disappointing. It has fallen especially short for researchers looking for clues to the genetic foundations of diseases that involve the interaction of multiple genes
“A growing body of evidence suggests that many, if not most, common diseases involve more than one gene and understanding how those genes interact is key to understanding the disease itself,” says Dr. Greenberg, a principal investigator in the Battelle Center for Mathematical Medicine
and corresponding author of the new study. “While it is possible to develop ameliorative therapeutics without that understanding, to effectively treat or even cure the disease, we must understand what causes the disease in the first place.”
The technique Dr. Greenberg’s team developed begins by using family data and focusing on alleles — the various forms of a single gene — at a location known to cause or be associated with a disease. By dividing the family data into groups, or strata, based on the presence of specific alleles, this new analysis method is better able to identify significant interaction between the disease gene and other genes in the genome. In this latest work, the scientists put the technique through a series of rigorous computer simulations designed to expose any weaknesses in the methodology. Based on the simulations, the power to detect interaction was 80 percent. The false positive was lower than 1 percent for the tests that were run.
While linkage analysis has been used in genomics for quite some time, the way the technology has been used has been limited, Dr. Greenberg says. By adding data stratification to the linkage analysis approach, Dr. Greenberg and his collaborators were able to not only identify loci involved in disease, but also pinpoint those genes’ interactions throughout the genome.
“While many disease-causing or associated alleles are known, to our knowledge there has been no linkage-based method to test for interactions of that known disease locus with other loci,” Dr. Greenberg says. “Our work gives investigators a way to not only test for interaction, but also to improve the ability to detect linkage in the first place. It’s one small step on the road to try to figure out how to find disease genes.”
The technique could be used in genetic screening of patients, he says, but it could also be used by researchers to identify the multiple genes in a gene-gene interacting disease. The researchers have used the technique in their studies of genetic linkages involved in familial primary pulmonary hypertension and are now using it in a study to better understand interactions between alleles involved in type 1 diabetes.