Editors' ChoiceHuman Genetics

Feeling SNP-y? Better Analysis Is the Cure

See allHide authors and affiliations

Science Translational Medicine  18 Jan 2012:
Vol. 4, Issue 117, pp. 117ec8
DOI: 10.1126/scitranslmed.3003694

Rapid advances in sequencing technologies are ushering in the era of personalized genomic medicine. Even those of us in the 99% can order a genome sequence, and, by 2014, a whole genome may be sequenced for $1000. The challenge then becomes not whether we can sequence every patient’s genome but how to interpret the explosion of genomic data. In particular, establishing the functional impact of rare genetic variants is a daunting task, requiring a savvy mix of bioinformatics, biochemistry, cell biology, and genetics. Now, Rees et al. have set the bar high in their systematic analysis of variants in GCKR.

The GCKR gene encodes the glucokinase regulatory protein, which plays a key role in glucose metabolism in the liver through competitive inhibition of hepatic glucokinase. Variants in GCKR have been associated with several adverse metabolic traits, including high serum triglycerides, elevated cholesterol, and diabetes risk. To discover the mode of action for rare variants in GCKR, the authors identified 19 protein-altering variants among 800 members of the ClinSeq study, a patient cohort enriched for cardiometabolic disease. Excluding one well-described common variant, all the variants were rare and correlated as a group, with higher serum triglycerides compared with wild-type controls. Because GCKR has been well studied, Rees and colleagues used an array of methods to characterize mutations in silico, in vitro, and in vivo. They generated reporter constructs for individual rare variants to assess each variant’s effect on protein localization, expression level, and ultimately inhibitory activity. When reevaluated in this way, rare variants fell into three groups of wild-type–like function, loss of function (LOF), or gain of function (GOF). Notably, the functional assessment showed that two variants were incorrectly classified by SIFT and PolyPhen, two popular algorithms for predicting mutation severity. Most importantly, reclassification of the variants based on their functional status revealed a significant correlation to elevations in triglycerides, low-density lipoprotein, and total cholesterol in the rare LOF variants and hinted at a possible protective effect of the GOF variants.

The work of Rees et al. highlights the contribution of rare genetic variants to common disease as well as the importance of a multidisciplinary approach in the functional analysis of genetic variants. Because phenotypic testing of all individual variants is impractical, improved prediction algorithms or newer disciplines like metabolomics will be critical in understanding a variant’s effects in disease and in guiding rational interventions to bring personal genomics to the bedside.

M. G. Rees et al., Correlation of rare coding variants in the gene encoding human glucokinase regulatory protein with phenotypic, cellular, and kinetic outcomes. J. Clin. Invest. 122, 205–217 (2012). [Abstract]

Stay Connected to Science Translational Medicine

Navigate This Article