The use of phenome-wide association studies (PheWAS) for exploration of novel genotype-phenotype relationships and pleiotropy discovery

Abstract

The field of phenomics has been investigating network structure among large arrays of phenotypes, and genome-wide association studies (GWAS) have been used to investigate the relationship between genetic variation and single diseases/outcomes. A novel approach has emerged combining both the exploration of phenotypic structure and genotypic variation, known as the phenome-wide association study (PheWAS). The Population Architecture using Genomics and Epidemiology (PAGE) network is a National Human Genome Research Institute (NHGRI)-supported collaboration of four groups accessing eight extensively characterized epidemiologic studies. The primary focus of PAGE is deep characterization of well-replicated GWAS variants and their relationships to various phenotypes and traits in diverse epidemiologic studies that include European Americans, African Americans, Mexican Americans/Hispanics, Asians/Pacific Islanders, and Native Americans. The rich phenotypic resources of PAGE studies provide a unique opportunity for PheWAS as each genotyped variant can be tested for an association with the wide array of phenotypic measurements available within the studies of PAGE, including prevalent and incident status for multiple common clinical conditions and risk factors, as well as clinical parameters and intermediate biomarkers. The results of PheWAS can be used to discover novel relationships between SNPs, phenotypes, and networks of interrelated phenotypes; identify pleiotropy; provide novel mechanistic insights; and foster hypothesis generation. The PAGE network has developed infrastructure to support and perform PheWAS in a high-throughput manner. As implementing the PheWAS approach has presented several challenges, the infrastructure and methodology, as well as insights gained in this project, are presented herein to benefit the larger scientific community.

Fig. 1

Workflow for PheWAS. Study samples are genotyped and quality control procedures are followed at each site independently. In addition all tests of association are calculated independently for each study. The PheWAS quality control step flags results as described in the text. As a resource for PAGE investigators, the most significant associations from PheWAS can be explored in the PAGE browser across race/ethnicity (green). For investigators outside of PAGE, aggregate data from PheWAS will be available via dbGAP. For the PAGE PheWAS analysis, results for six studies will be used and where SNPs and phenotypes exist across data sets, replication will be sought (blue). Significant results can then be selectively investigated and characterized further.

Fig. 2

Skewness and transformation. Skewness measured for 351 phenotypes, stratified by race/ethnicity, before transformation and after adding 1 then natural log transforming each continuous variable. Results are presented in the upper and lower panel for: African American (AA) untransformed, AA transformed, European American (EA) untransformed, EA transformed, Hispanic (H) untransformed, H transformed, Asian Pacific Islander (API) transformed, API untransformed, American Indians (AI) untransformed, AI transformed. Horizontal red lines are at positive two and negative two. The lower panel shows a “zoomed in” version of the upper panel, showing the compression of points after the transformation of variables.

Fig. 2

Skewness and transformation. Skewness measured for 351 phenotypes, stratified by race/ethnicity, before transformation and after adding 1 then natural log transforming each continuous variable. Results are presented in the upper and lower panel for: African American (AA) untransformed, AA transformed, European American (EA) untransformed, EA transformed, Hispanic (H) untransformed, H transformed, Asian Pacific Islander (API) transformed, API untransformed, American Indians (AI) untransformed, AI transformed. Horizontal red lines are at positive two and negative two. The lower panel shows a “zoomed in” version of the upper panel, showing the compression of points after the transformation of variables.

Fig. 3

PAGE browser. Different views of the PAGE browser provide different possibilities for investigators examining the results of the analyses. (A) The heat map view, where association results are plotted in a “heat map” format. SNPs are along x-axis, phenotypes are along the y-axis. The color of each cell corresponding to the individual SNP-by-Phenotype results is either a shade of yellow or red if the direction of effect is positive, or a shade of green or blue if the direction of the effect is negative. Increasing significance is indicated by shade respective to the direction of effect. (B) Phenotypic summary information across race/ethnicity. The box-plots show the 1st, 2nd, and 3rd quantiles by genotype for quantitative phenotypes. (C) For associations, forest plots are available, stratified across race/ethnicity.

Fig. 3

PAGE browser. Different views of the PAGE browser provide different possibilities for investigators examining the results of the analyses. (A) The heat map view, where association results are plotted in a “heat map” format. SNPs are along x-axis, phenotypes are along the y-axis. The color of each cell corresponding to the individual SNP-by-Phenotype results is either a shade of yellow or red if the direction of effect is positive, or a shade of green or blue if the direction of the effect is negative. Increasing significance is indicated by shade respective to the direction of effect. (B) Phenotypic summary information across race/ethnicity. The box-plots show the 1st, 2nd, and 3rd quantiles by genotype for quantitative phenotypes. (C) For associations, forest plots are available, stratified across race/ethnicity.

Fig. 3

PAGE browser. Different views of the PAGE browser provide different possibilities for investigators examining the results of the analyses. (A) The heat map view, where association results are plotted in a “heat map” format. SNPs are along x-axis, phenotypes are along the y-axis. The color of each cell corresponding to the individual SNP-by-Phenotype results is either a shade of yellow or red if the direction of effect is positive, or a shade of green or blue if the direction of the effect is negative. Increasing significance is indicated by shade respective to the direction of effect. (B) Phenotypic summary information across race/ethnicity. The box-plots show the 1st, 2nd, and 3rd quantiles by genotype for quantitative phenotypes. (C) For associations, forest plots are available, stratified across race/ethnicity.

Fig. 4

Approaches for visualizing and interpreting results of PheWAS. There are multiple ways the results of PheWAS can be explored. Correlations between quantitative phenotypes can be calculated to investigate some of the relationships between phenotypes. Results across race/ethnicity can be compared. Results can be ranked by significance and then explored one at a time for further replication and characterization with further phenotype harmonization. The expected/unexpected nature of results can also be investigated, as some genotype-phenotype associations are previously known. Grouping of phenotypes is possible, as well as further exploration of phenotypic structure.

Fig. 5

Correlation Heat Map from EAGLE phenotypic and PheWAS data. Phenotypes are along the x-axis. The middle track shows the −log10(P-value) results for each genotype-phenotype test of association. Below the P-value significance track, a heat map of correlation results is plotted. Pearson correlations between quantitative phenotypes are calculated and the absolute value of each correlation is plotted. Relative significance is displayed by the shade of the color (that is, the more significant the correlation, the brighter yellow the cell). In this figure, bone mineral content and bone mineral density measurements are correlated, along with body mass index.