Genome-Wide Association Studies and Genomic Prediction

Genome-wide association studies (GWAS) have rapidly spread across the globe over the last few years becoming the de facto approach to identify candidate regions associated with complex diseases in human medicine. GWAS and genome analysis are also powerful tools to provide a handle on genetic variation in a wide range of traits important for human health, agriculture, conservation, and the evolutionary history of life. As the technology matured and results from the various studies came to light, a clear picture emerged that, as often seems to be the case, biological processes prove to be much more complex than we would want them to be. While many new insights were and are being attained, maybe just as many new questions emerged. In many cases, GWAS successfully identified variants that were unquestionably associated with variation in traits and, in some cases, culminated in the discovery of the causal variant(s) and its mechanism of action. In other cases, these studies evidenced that a large number of traits are highly polygenic, and, instead of identifying a few genomic region 'culprits', we saw a scattering of effects across the whole genome. While in these latter cases the underlying biology largely remains elusive, it still allows making predictions of phenotypes based on the genomic information of an individual. And this in the short term might be even more important for translational outcomes. This brings us to the focus of this volume in the series 'Methods in Molecular Biology: Genome-Wide Association Studies and Genomic Prediction'. This volume in the series covers from the preliminary stages of understanding the phenotypes of interest and design issues for GWAS, passing through efficient computational methods to store and handle large datasets, quality control measures, phasing, haplotype inference and imputation, and moving on to then discuss the various statistical approaches to data analysis where the experimental objective is either to nail down the biology by identifying genomic regions associated to a trait or to use the data to make genomic predictions about a future phenotypic outcome (e.g. predict onset of disease).