Phenome-Wide Association Studies: Embracing Complexity for Discovery

Genetic variation in ALDH4A1 is associated with muscle health over the lifespan and across species

The influence of genetic variation on the aging process, including the incidence and severity of age-related diseases, is complex. Here, we define the evolutionarily conserved mitochondrial enzyme ALH-6/ALDH4A1 as a predictive biomarker for age-related changes in muscle health by combining Caenorhabditis elegans genetics and a gene-wide association scanning (GeneWAS) from older human participants of the US Health and Retirement Study (HRS). In a screen for mutations that activate oxidative stress responses, specifically in the muscle of C. elegans, we identified 96 independent genetic mutants harboring loss-of-function alleles of alh-6, exclusively. Each of these genetic mutations mapped to the ALH-6 polypeptide and led to the age-dependent loss of muscle health. Intriguingly, genetic variants in ALDH4A1 show associations with age-related muscle-related function in humans. Taken together, our work uncovers mitochondrial alh-6/ALDH4A1 as a critical component to impact normal muscle aging across species and a predictive biomarker for muscle health over the lifespan.

Ageing is inevitable, but what makes one person ‘age well’ and another decline more quickly remains largely unknown. While many aspects of ageing are clearly linked to genetics, the specific genes involved often remain unidentified.

Sarcopenia is an age-related condition affecting the muscles. It involves a gradual loss of muscle mass that becomes faster with age, and is associated with loss of mobility, decreased quality of life, and increased risk of death. Around half of all people aged 80 and over suffer from sarcopenia. Several lifestyle factors, especially poor diet and lack of exercise, are associated with the condition, but genetics is also involved: the condition accelerates more quickly in some people than others, and even fit, physically active individuals can be affected.

To study the genetics of conditions like sarcopenia, researchers often use animals like flies or worms, which have short generation times but share genetic similarities with humans. For example, the worm Caenorhabditis elegans has equivalents of several human muscle genes, including the gene alh-6. In worms, alh-6 is important for maintaining energy supply to the muscles, and mutating it not only leads to muscle damage but also to premature ageing. Given this insight, Villa, Stuhr, Yen et al. wanted to determine if variation in the human version of alh-6, ALDH4A1, also contributes to individual differences in muscle ageing and decline in humans.

Evaluating variation in this gene required a large amount of genetic data from older adults. These were taken from a continuous study that follows >35,000 older adults. Importantly, the study collects not only information on gene sequences but also measures of muscle health and performance over time for each individual. Analysis of these genetic data revealed specific small variations in the DNA of ALDH4A1, all of which associated with reduced muscle health.

Follow-up experiments in worms used genetic engineering techniques to test how variation in the worm alh-6 gene could influence age-related health. The resulting mutant worms developed muscle problems much earlier than their normal counterparts, supporting the role of alh-6/ALDH4A1 in determining muscle health across the lifespan of both worms and humans.

These results have identified a key influencer of muscle health during ageing in worms, and emphasize the importance of validating effects of genetic variation among humans during this process. Villa, Stuhr, Yen et al. hope that this study will help researchers find more genetic ‘markers’ of muscle health, and ultimately allow us to predict an individual’s risk of sarcopenia based on their genetic make-up.

Quantitative analyses of adiposity dynamics in zebrafish

Adipose tissues often exhibit subtle, quantitative differences between individuals, leading to a graded series of adiposity phenotypes at the population level. Robust, quantitative analyses are vital for studying these differences. In this Commentary we highlight two articles from our lab that employ sensitive new methods in zebrafish capable of delineating complex and quantitative adiposity phenotypes. In the first article, we utilized in vivo imaging to systematically quantify zebrafish adipose tissues. We identified 34 regionally distinct zebrafish adipose tissues and developed statistical models to predict the size and variance of each adipose tissue over the course of zebrafish growth. We then employed these models to identify effects of strain and diet on adipose tissue growth. In the second article, we employed deep phenotyping to study complex disease-related adiposity traits. Using this methodology, we identified that adipose tissues have unique capacities to re-deposit lipid following food restriction and re-feeding. These distinct re-deposition potentials led to widespread fat distribution changes following re-feeding. We discuss how these novel findings may provide relevance to health conditions such as anorexia nervosa. Together, the strategies described in these two articles can be used as unbiased and quantitative methods to uncover new relationships between genotype, diet and adiposity.

Multi-Phenotype Association Decomposition: Unraveling Complex Gene-Phenotype Relationships

Various patterns of multi-phenotype associations (MPAs) exist in the results of Genome Wide Association Studies (GWAS) involving different topologies of single nucleotide polymorphism (SNP)-phenotype associations. These can provide interesting information about the different impacts of a gene on closely related phenotypes or disparate phenotypes (pleiotropy). In this work we present MPA Decomposition, a new network-based approach which decomposes the results of a multi-phenotype GWAS study into three bipartite networks, which, when used together, unravel the multi-phenotype signatures of genes on a genome-wide scale. The decomposition involves the construction of a phenotype powerset space, and subsequent mapping of genes into this new space. Clustering of genes in this powerset space groups genes based on their detailed MPA signatures. We show that this method allows us to find multiple different MPA and pleiotropic signatures within individual genes and to classify and cluster genes based on these SNP-phenotype association topologies. We demonstrate the use of this approach on a GWAS analysis of a large population of 882 Populus trichocarpa genotypes using untargeted metabolomics phenotypes. This method should prove invaluable in the interpretation of large GWAS datasets and aid in future synthetic biology efforts designed to optimize phenotypes of interest.

Genomic and Phenomic Research in the 21st Century

The field of human genomics has changed dramatically over time. Initial genomic studies were predominantly restricted to rare disorders in small families. Over the past decade, researchers changed course from family-based studies and instead focused on common diseases and traits in populations of unrelated individuals. With further advancements in biobanking, computer science, electronic health record (EHR) data, and more affordable high-throughput genomics, we are experiencing a new paradigm in human genomic research. Rapidly changing technologies and resources now make it possible to study thousands of diseases simultaneously at the genomic level. This review will focus on these advancements as scientists begin to incorporate phenome-wide strategies in human genomic research to understand the etiology of human diseases and develop new drugs to treat them.

Rare variants in drug target genes contributing to complex diseases, phenome-wide

The DrugBank database consists of ~800 genes that are well characterized drug targets. This list of genes is a useful resource for association testing. For example, loss of function (LOF) genetic variation has the potential to mimic the effect of drugs, and high impact variation in these genes can impact downstream traits. Identifying novel associations between genetic variation in these genes and a range of diseases can also uncover new uses for the drugs that target these genes. Phenome Wide Association Studies (PheWAS) have been successful in identifying genetic associations across hundreds of thousands of diseases. We have conducted a novel gene based PheWAS to test the effect of rare variants in DrugBank genes, evaluating associations between these genes and more than 500 quantitative and dichotomous phenotypes. We used whole exome sequencing data from 38,568 samples in Geisinger MyCode Community Health Initiative. We evaluated the results of this study when binning rare variants using various filters based on potential functional impact. We identified multiple novel associations, and the majority of the significant associations were driven by functionally annotated variation. Overall, this study provides a sweeping exploration of rare variant associations within functionally relevant genes across a wide range of diagnoses.

Embracing Complex Associations in Common Traits: Critical Considerations for Precision Medicine

Genome-wide association studies (GWAS) have identified numerous loci associated with human phenotypes. This approach, however, does not consider the richly diverse and complex environment with which humans interact throughout the life course, nor does it allow for interrelationships between genetic loci and across traits. As we move toward making precision medicine a reality, whereby we make predictions about disease risk based on genomic profiles, we need to identify improved predictive models of the relationship between genome and phenome. Methods that embrace pleiotropy (the effect of one locus on more than one trait), and gene-environment (G×E) and gene-gene (G×G) interactions, will further unveil the impact of alterations in biological pathways and identify genes that are only involved with disease in the context of the environment. This valuable information can be used to assess personal risk and choose the most appropriate medical interventions based on the genotype and environment of an individual, the whole premise of precision medicine.

Multiphenotype association study of patients randomized to initiate antiretroviral regimens in AIDS Clinical Trials Group protocol A5202

Supplemental Digital Content is available in the text.

Background

High-throughput approaches are increasingly being used to identify genetic associations across multiple phenotypes simultaneously. Here, we describe a pilot analysis that considered multiple on-treatment laboratory phenotypes from antiretroviral therapy-naive patients who were randomized to initiate antiretroviral regimens in a prospective clinical trial, AIDS Clinical Trials Group protocol A5202.

Participants and methods

From among 5 9545 294 polymorphisms imputed genome-wide, we analyzed 2544, including 2124 annotated in the PharmGKB, and 420 previously associated with traits in the GWAS Catalog. We derived 774 phenotypes on the basis of context from six variables: plasma atazanavir (ATV) pharmacokinetics, plasma efavirenz (EFV) pharmacokinetics, change in the CD4+ T-cell count, HIV-1 RNA suppression, fasting low-density lipoprotein-cholesterol, and fasting triglycerides. Permutation testing assessed the likelihood of associations being by chance alone. Pleiotropy was assessed for polymorphisms with the lowest P-values.

Results

This analysis included 1181 patients. At P less than 1.5×10⁻⁴, most associations were not by chance alone. Polymorphisms with the lowest P-values for EFV pharmacokinetics (CYPB26 rs3745274), low-density lipoprotein -cholesterol (APOE rs7412), and triglyceride (APOA5 rs651821) phenotypes had been associated previously with those traits in previous studies. The association between triglycerides and rs651821 was present with ATV-containing regimens, but not with EFV-containing regimens. Polymorphisms with the lowest P-values for ATV pharmacokinetics, CD4 T-cell count, and HIV-1 RNA phenotypes had not been reported previously to be associated with that trait.

Conclusion

Using data from a prospective HIV clinical trial, we identified expected genetic associations, potentially novel associations, and at least one context-dependent association. This study supports high-throughput strategies that simultaneously explore multiple phenotypes from clinical trials’ datasets for genetic associations.

An Analytic Solution to the Computation of Power and Sample Size for Genetic Association Studies under a Pleiotropic Mode of Inheritance

Our motivation here is to calculate the power of 3 statistical tests used when there are genetic traits that operate under a pleiotropic mode of inheritance and when qualitative phenotypes are defined by use of thresholds for the multiple quantitative phenotypes. Specifically, we formulate a multivariate function that provides the probability that an individual has a vector of specific quantitative trait values conditional on having a risk locus genotype, and we apply thresholds to define qualitative phenotypes (affected, unaffected) and compute penetrances and conditional genotype frequencies based on the multivariate function. We extend the analytic power and minimum-sample-size-necessary (MSSN) formulas for 2 categorical data-based tests (genotype, linear trend test [LTT]) of genetic association to the pleiotropic model. We further compare the MSSN of the genotype test and the LTT with that of a multivariate ANOVA (Pillai). We approximate the MSSN for statistics by linear models using a factorial design and ANOVA. With ANOVA decomposition, we determine which factors most significantly change the power/MSSN for all statistics. Finally, we determine which test statistics have the smallest MSSN. In this work, MSSN calculations are for 2 traits (bivariate distributions) only (for illustrative purposes). We note that the calculations may be extended to address any number of traits. Our key findings are that the genotype test usually has lower MSSN requirements than the LTT. More inclusive thresholds (top/bottom 25% vs. top/bottom 10%) have higher sample size requirements. The Pillai test has a much larger MSSN than both the genotype test and the LTT, as a result of sample selection. With these formulas, researchers can specify how many subjects they must collect to localize genes for pleiotropic phenotypes.

Evidence for extensive pleiotropy among pharmacogenes

AIM

We sought to identify potential pleiotropy involving pharmacogenes.

METHODS

We tested 184 functional variants in 34 pharmacogenes for associations using a custom grouping of International Classification and Disease, Ninth Revision billing codes extracted from deidentified electronic health records of 6892 patients.

RESULTS

We replicated several associations including ABCG2 (rs2231142) and gout (p = 1.73 × 10(-7); odds ratio [OR]: 1.73; 95% CI: 1.40-2.12); and SLCO1B1 (rs4149056) and jaundice (p = 2.50 × 10(-4); OR: 1.67; 95% CI: 1.27-2.20).

CONCLUSION

In this systematic screen for phenotypic associations with functional variants, several novel genotype-phenotype combinations also achieved phenome-wide significance, including SLC15A2 rs1143672 and renal osteodystrophy (p = 2.67 × 10(-) (6); OR: 0.61; 95% CI: 0.49-0.75).

From Genome-Wide Association Study to Phenome-Wide Association Study: New Paradigms in Obesity Research

Obesity is a condition in which excess body fat has accumulated over an extent that increases the risk of many chronic diseases. The current clinical classification of obesity is based on measurement of body mass index (BMI), waist-hip ratio, and body fat percentage. However, these measurements do not account for the wide individual variations in fat distribution, degree of fatness or health risks, and genetic variants identified in the genome-wide association studies (GWAS). In this review, we will address this important issue with the introduction of phenome, phenomics, and phenome-wide association study (PheWAS). We will discuss the new paradigm shift from GWAS to PheWAS in obesity research. In the era of precision medicine, phenomics and PheWAS provide the required approaches to better definition and classification of obesity according to the association of obese phenome with their unique molecular makeup, lifestyle, and environmental impact.

Unravelling the human genome-phenome relationship using phenome-wide association studies

Advances in genotyping technology have, over the past decade, enabled the focused search for common genetic variation associated with human diseases and traits. With the recently increased availability of detailed phenotypic data from electronic health records and epidemiological studies, the impact of one or more genetic variants on the phenome is starting to be characterized both in clinical and population-based settings using phenome-wide association studies (PheWAS). These studies reveal a number of challenges that will need to be overcome to unlock the full potential of PheWAS for the characterization of the complex human genome-phenome relationship.