New tool for genome-wide associati...( Modern genotyping technologies offer...)

Modern genotyping technologies offer new opportunities to explore how genes influence health and disease, but also present the challenge of analyzing huge amounts of genetic and clinical data. With this in mind, investigators at the Lausanne Branch of the international Ludwig Institute for Cancer Research (LICR), the Swiss Institute of Bioinformatics (SIB) and the University Hospital of Lausanne have developed AssociationViewer, a computational tool that displays genetic differences between individuals on a large scale. Presented in the March 1st (Vol. 25/5) issue of Bioinformatics, the software is a public resource that will help scientists and physicians discover new genetic markers for diseases and other conditions.

Less than 0.5% of the human genome differs between individuals, but this small fraction holds clues that can reveal how each person will respond to a particular disease, therapy or environmental factor. In the past years, genome-wide association (GWA) studiescomprehensive surveys that look for differences in people's DNA codehave uncovered millions of small genetic variations known as single-nucleotide polymorphisms (SNPs). By sharing GWA data, scientists have gained insight into the genetic basis of many complex diseases including cancer and heart disease, and embarked on developing personalized therapeutics.

According to Dr. Nicolas Guex, a senior bioinformatician at the SIB, those who carry out GWA studies need simple ways to integrate datasets from different sources. They also rely on tools that help them visualize SNPs across the entire map of the genome. "AssociationViewer provides an elegant solution to efficiently mine the wealth of data to generate hypotheses," he says. Another user, Dr. Sam Deutsch at the U.S. Department of Energy Joint Genome Institute, USA, also considers the software to be a powerful tool for computational genetics. "Compared to other tools out there, AssociationViewer brings more flexible
'/>"/>