Population Genetics and Pharmacogenetics

Abstract

A clear indication of the increasing prominence of population genetics in pediatric endocrinology is that these papers are becoming a regular feature across the other chapters of the Yearbook. For example, in other chapters this year look out for the summaries of major genetic association studies of type 1 diabetes [1][2][3], blood glucose levels [4,5] and bone mineral density [6], and also for an example of the emerging role of pharmacogenetics in a study of a vitamin D receptor genotype [7]. In the last year, genome-wide association (GWA) studies have continued to lead on new discoveries of single nucleotide polymorphisms (SNPs) in genes or chromosomal positions (loci) that had previously not been suspected of acting in the various disease or physiological pathways. Indeed, the pace of discovery has increased with the pooling of existing data into large international consortia for meta-analyses; type 1 diabetes [1], obesity [see Willer et al. below], and blood pressure [8] are examples of such highly fruitful collaborations. So what uses do these genetic discoveries have? There is widespread acceptance that the effect sizes of individual SNPs are much lower than were previously thought. For example, an odds ratio of >1.3 is now considered to be a large genetic effect, and nearly all of the 18 new SNPs for type 1 diabetes have odds ratios of <1.2 [1]. This means that such variants do not have, and possibly may never have, sufficient power to predict future disease risks in individuals [see Lyssenko et al. and Meigs et al. below]. New studies looking at using copy number variation or low-frequency polymorphisms may identify loci with larger effect sizes and greater ability to predict disease risks. The current discoveries have started to inform new animal models of disease, such as the Fto knockout mouse [9] [see chapter on Obesity and Weight Regulation], but it is far too early to expect any impact on drug discovery. For the time being, recognition of new loci for common traits (e.g. age at menarche) will hopefully inform the identification of rare mutations or chromosomal microdeletions that cause severe clinical conditions (e.g. precocious puberty or hypogonadotrophic hypogonadism). In combination with international clinical collaborations, this could inform genetic counseling for affected families [see Firth et al. below]. Secondly, by Mendelian randomization principles the use of these robust variants as 'instrumental variables' to demonstrate causal effects between exposures and disease outcomes is a very powerful new epidemiological tool [see comments on Sulem et al., Perry et al., Liu et al. and Eriksson et al. below]. Science 2009;324:387-389 Background: Genome-wide association (GWA) studies have identified genomic regions contributing to common human diseases, but they often do not identify the precise causative genes and sequence variants. Methods: In order to identify specific causative variants for type 1 diabetes (T1D), the authors resequenced exons and splice sites of 10 candidate genes in ...