Gene mapping: Fishing for LD in human populations

Jane Alfred, Associate Editor, Nature Reviews Genetics

Fishing for genes in specific populations is a common pastime among human geneticists. This is because genetically isolated populations - those that have been through a bottleneck and then rapid expansion - have provided rich fishing grounds for researchers in search of disease genes that cause rare, monogenic disorders. In such populations, including those of Finland and Sardinia, single-origin founder mutations are present in many people, making a disease more common than in an outbred population. The genetic heterogeneity of a disease is also reduced in an isolated population, and so disease genes are easier to locate and identify.

The bait that is used to fish out founder mutations from an isolated gene pool is linkage disequilibrium (LD). LD describes the tendency of genetic loci to be inherited together more often than would be expected if they had randomly segregated in a population. It is thought to reflect both a populationÕs history and the physical distance between genetic markers. The strong association of markers with disease in isolated populations has allowed LD mapping to be successfully used to identify monogenic disease genes in these populations. It has therefore been assumed that this approach will also help to identify loci that underlie common, and more genetically complex, diseases. In this monthÕs Nature Genetics, however, two papers raise some questions about this assumption.

These two LD analyses of people from Sardinia, Finland and of mixed descent gave a surprising result: the genetically isolated populations showed similar levels of LD to the outbred ones. Patricia Taillon-Miller et al. typed 39 markers from the Xq25-Xq28 region in three populations samples comprising 100 Finns, 150 Sardinians and a mixed-European sample of 92 people. Their analysis found two regions of strong LD on the X chromosome, the extent and location of which was similar in all three samples. Iain Eaves et al. found similar mean levels of LD in population samples from Finland, Sardinia, the UK and the USA in an analysis of 21 markers from chromosome 18. Both studies found that LD levels could not be explained by differences in between-marker distances or allele frequency alone.

So what implications do these results have for gene hunters who are looking for the best fishing grounds? The answer appears to depend on the type of disease gene you are looking for. Eaves et al. suggest large, non-founder populations should be included in the search for variants that underlie common disease, and advocate collecting clinical material from these populations. This argument would seem reasonable on the basis of the ecommon disease-common varianti hypothesis (which proposes that common diseases are caused by the segregation of old, common variants). Indeed, the use of population simulations to estimate the extent of LD surrounding common gene variants in the human population supports these data. However, Michael Boenhke, in an accompanying News and Views article, warns against overgeneralizing these results given that only two chromosomal regions in a few populations of predominantly European descent were studied. Future insight into this complex area might warrant a delay of that fishing trip until the fishing grounds are better charted (with a whole-genome LD map) or perhaps a change of bait.

Reprinted with permission from Nature Reviews Genetics (http://www.nature.com/reviews/genetics)

Original research papers

Eaves, I. A. et al. The genetically isolated populations of Finland and Sardinia may not be a panacea for linkage disequilibrium mapping of common disease genes. Nature Genet. 25, 320-323 (2000).

Taillon-Miller, P. et al. Juxtaposed regions of extensive and minimal linkage disequilibrium in human Xq25 and Xq28. Nature Genet. 25, 324-328 (2000).