Structure and Consequences of Human Genome Variation
Tony Brookes, Center for Genomics and Bioinformatics, Karolinska Institute

1) Improved Methods for DNA Variation Analysis Today’s systems for single nucleotide polymorphism (SNP) genotyping are impressive, but inadequate for whole human genome analyses. Furthermore, we recently discovered that extensive genome regions vary in copy number, making 'conventional' SNP genotyping data within such regions invalid. To solve the problems, we are working towards a) an ultra-high-throughput and flexible SNP genotyping system based upon DASH (Dynamic Allele-Specific Hybridization), b) a means to amplify an unlimited number of target fragments from the genome for parallel analysis, and c) a means to score genome copy number variation in a high-throughput manner. 2) Bioinformatics Aspects of DNA Variation Analysis Our bioinformatics activities in support of genome variation research concern tools for automated assay design, automated interpretation of analysis device outputs, genotype data handling, and result analysis. One major component of this is the establishment of a comprehensive public-domain database summarizing all that is known about disease phenotypes and their underlying genotype associations. For this, we have devised an elegant solution that allows any form of phenotype to be represented in a standardized data structure. 3) Population & Evolutionary Genetics Using tools developed by us and others, we are exploring basic questions about the structure of human DNA variation. One particular interest is our recent breakthrough discovery that large stretches of DNA are non-unique and vary in presence/absence between individuals. These and other types of sequence variation are being investigated for their distribution in various global samples, to help understand the forces of evolution and human population history. 4) Disease Risk and Drug Response To study the biological consequences of DNA variation, we are using extensive clinical materials collected over decades by researchers in Sweden (where Prof Brookes holds an adjunct academic position). These large, well-phenotyped and ideally homogenous materials allow us to explore the subtleties of how genome variation impacts disease. Alzheimer’s Disease, Rheumatoid Arthritis, Diabetes, and Cardiovascular Disease are major areas of interest. We are working towards a situation in which a sufficiently large fraction of the genetic risk of disease is understood, so that predictive and personalized