Principles Of Mendelian Randomization Research Articles

Learning Causal Biological Networks With the Principle of Mendelian Randomization.

Although large amounts of genomic data are available, it remains a challenge to reliably infer causal (i. e., regulatory) relationships among molecular phenotypes (such as gene expression), especially when multiple phenotypes are involved. We extend the interpretation of the Principle of Mendelian randomization (PMR) and present MRPC, a novel machine learning algorithm that incorporates the PMR in the PC algorithm, a classical algorithm for learning causal graphs in computer science. MRPC learns a causal biological network efficiently and robustly from integrating individual-level genotype and molecular phenotype data, in which directed edges indicate causal directions. We demonstrate through simulation that MRPC outperforms several popular general-purpose network inference methods and PMR-based methods. We apply MRPC to distinguish direct and indirect targets among multiple genes associated with expression quantitative trait loci. Our method is implemented in the R package MRPC, available on CRAN (https://cran.r-project.org/web/packages/MRPC/index.html).

Abstract P096: The Genetics Architecture of the Serum Metabolome

Introduction: The metabolome is a collection of small molecules in a biologic sample, and may serve as biomarkers or predictors of heart disease. Whole genome sequence analysis offers the opportunity to investigate rare and low-frequency annotated variants across the human genome. We used whole genome sequence analysis to characterize the genetic architecture of the serum metabolome. Methods: Whole genome sequencing and measurement (chromotagraphy and mass spectroscopy) of 245 serum metabolites were done in 1,458 European Americans and 1,679 African Americans from the Atherosclerosis Risk in Communities (ARIC) study, and these data were used to perform a trans-ethnic meta-analysis. Common variants (MAF>5%) were analyzed individually using an additive genetic model. Rare and low-frequency protein-altering variants (MAF≤5%) were aggregated by genes. In order to determine the contribution of regulatory and non-protein coding regions of the genome, we conducted aggregate tests across the entire genome using a 4kb sliding window as well as in predefined regulatory elements, which includs enhancers, promoter, and 3’ and 5’ untranslated region of a gene. Results: We identified 119 significant associations between genetic variants and metabolite levels (significance threshold p<2.0*10 -10 for single variants, p<2.9х10 -10 for aggregate tests), of which 49 were novel, including genes involved in known Mendelian conditions, protein biological processes, and disease related pathways. Six genes ( DMGDH, AGA, ACY1, PRODH, DDC and CPS1 ) causing rare inborn errors of metabolism were associated with amino acid levels in the general population. A predicated regulatory variant in the AGA gene, encoding a protein involved in asparagine generation, was associated with serum asparagine levels independent of any coding variants in this gene. Seven genes ( ABCC2, PKD2L1, SLC10A1, FDX1, CYP3A43, UGT2B15 and SULT2A1 ) related to lipid-related metabolite levels were identified, whose gene products are involved in secretion, channeling and transportation. Analysis of regulatory regions unraveled associations between three steroid lipids and a member of the cytochrome P450 family, CYP3A43 . Five genes within the kinin-kallikrein pathway were identified to be related to small peptide levels, including KLKB1 , KNG1 , F12 , ACE and CPN1 . Variants in CPN1 , which is known to bind to fibrinogen, were associated with DSGEGDFXAEGGGVR, a peptide which is produced during fibrinogen to fibrin conversion. Conclusion: This study outlines an approach to characterize the genetic architecture of the human serum metabolome and shows that sequence variants affect multiple human metabolites. Using the principle of Mendelian randomization, the next step is to determine whether any of these metabolites are in causal pathways to disease.

Loss-of-function variants influence the human serum metabolome.

The metabolome is a collection of small molecules resulting from multiple cellular and biological processes that can act as biomarkers of disease, and African-Americans exhibit high levels of genetic diversity. Exome sequencing of a sample of deeply phenotyped African-Americans allowed us to analyze the effects of annotated loss-of-function (LoF) mutations on 308 serum metabolites measured by untargeted liquid and gas chromatography coupled with mass spectrometry. In an independent sample, we identified and replicated four genes harboring six LoF mutations that significantly affected five metabolites. These sites were related to a 19 to 45% difference in geometric mean metabolite levels, with an average effect size of 25%. We show that some of the affected metabolites are risk predictors or diagnostic biomarkers of disease and, using the principle of Mendelian randomization, are in the causal pathway of disease. For example, LoF mutations in SLCO1B1 elevate the levels of hexadecanedioate, a fatty acid significantly associated with increased blood pressure levels and risk of incident heart failure in both African-Americans and an independent sample of European-Americans. We show that SLCO1B1 LoF mutations significantly increase the risk of incident heart failure, thus implicating the metabolite in the causal pathway of disease. These results reveal new avenues into gene function and the understanding of disease etiology by integrating -omic technologies into a deeply phenotyped population study.

Association of a Fasting Glucose Genetic Risk Score With Subclinical Atherosclerosis

OBJECTIVEElevated fasting glucose level is associated with increased carotid intima-media thickness (IMT), a measure of subclinical atherosclerosis. It is unclear if this association is causal. Using the principle of Mendelian randomization, we sought to explore the causal association between circulating glucose and IMT by examining the association of a genetic risk score with IMT.RESEARCH DESIGN AND METHODSThe sample was drawn from the Atherosclerosis Risk in Communities (ARIC) study and included 7,260 nondiabetic Caucasian individuals with IMT measurements and relevant genotyping. Components of the fasting glucose genetic risk score (FGGRS) were selected from a fasting glucose genome-wide association study in ARIC. The score was created by combining five single nucleotide polymorphisms (SNPs) (rs780094 [GCKR], rs560887 [G6PC2], rs4607517 [GCK], rs13266634 [SLC30A8], and rs10830963 [MTNR1B]) and weighting each SNP by its strength of association with fasting glucose. IMT was measured through bilateral carotid ultrasound. Mean IMT was regressed on the FGGRS and on the component SNPs, individually.RESULTSThe FGGRS was significantly associated (P = 0.009) with mean IMT. The difference in IMT predicted by a 1 SD increment in the FGGRS (0.0048 mm) was not clinically relevant but was larger than would have been predicted based on observed associations between the FFGRS, fasting glucose, and IMT. Additional adjustment for baseline measured glucose in regression models attenuated the association by about one third.CONCLUSIONSThe significant association of the FGGRS with IMT suggests a possible causal association of elevated fasting glucose with atherosclerosis, although it may be that these loci influence IMT through nonglucose pathways.

Exploring a Cancer Biomarker: The Example of C-Reactive Protein

An elevated plasma level of C-reactive protein (CRP) is an established biomarker of increased risk of cardiovascular diseases and diabetes ( 1 , 2 ). An association has also been suggested between plasma CRP level and the risk of cancer, particularly lung and colon cancers; however, the evidence is inconclusive because of the lack of large-scale prospective studies in which CRP levels were measured in healthy individuals before the onset of the disease ( 3 ). One explanation for why CRP may identify individuals at increased risk of disease lies in its association with chronic — mostly subclinical — inflammation. There is also some evidence that CRP may act as active contributor to inflammation, and this activity may depend on the conformation of the protein ( 4 ). Whether CRP is only a biomarker or an active player in inflammation might have important clinical implications. The study by Allin et al. ( 5 ) in this issue of the Journal addresses precisely the role of CRP as a biomarker vs as a contributor to carcinogenesis. The study is an elegant example of how genetic variants that have a functional impact can be used to explore associations between environmental factors and disease, and specifi cally to identify and control for confounding factors, based on the approach that has become known as Mendelian randomization, a term fi rst used by Gray and Wheatley in 1991 in the context of treatment of childhood leukemia ( 6 ). The basic principle of Mendelian randomization is that if a genetic variant infl uences a nongenetic (ie, environmental) factor that is relevant to disease risk, and if the same variant has no other associations with the same disease, then the variant predicts disease risk through its infl uence on the risk factor. Therefore, genetic variants with well-characterized functions (or that are in linkage disequilibrium with other variants with a well-characterized function) can be used to characterize associations between environmental factors and disease risk ( 7 ). Furthermore, using genetic variants instead of their environmental counterparts ensures protection from confounding and reverse causality and — at least in some cases — allows one to assess the effect of long-term exposures. The requirements for Mendelian randomization to be useful (strong functional signifi cance of the variant and the absence of an effect on other pathways) are demanding and are not often fulfi lled. There are examples, however, in which the analysis of genetic variants has helped to elucidate the role of environmental exposures, such as polymorphisms in genes encoding alcoholmetabolizing enzymes that explain the possible protective role of alcohol intake in cardiovascular disease as well as its detrimental role in pregnancy outcomes ( 8 , 9 ). The Mendelian randomization approach would be particularly valuable in the case of environmental factors that are not easily measurable.