Perils in the use of linkage disequilibrium for fine gene mapping: simple insights from population genetics

It is generally believed that genome-wide association (GWA) studies stand a good chance for finding susceptibility genes for common complex diseases. Although the results thus far have been somewhat promising, there are still many inherent difficulties and many initial associations do not get replicated. The common strategy in GWA studies has been that of selecting the most statistically significant single nucleotide polymorphisms with the hope that these will be very physically close to causal variants because of strong linkage disequilibrium (LD). Using simple ideas from population genetics, this commentary explains why this strategy can be misleading. It argues that there is an intrinsic problem in the way LD is currently used for fine-mapping. This is because most of the metrics that are currently used to measure LD are inadequate, as they do not take into account evolutionary variables that shape the LD structure of the human genome. Recent research on another metric, based on Malécot's model for isolation by distance, holds considerable promise for GWA studies and merits more serious consideration by geneticists.