Inferring the direction of a causal link and estimating its effect via a Bayesian Mendelian randomization approach
- PMID: 31146640
- PMCID: PMC7221461
- DOI: 10.1177/0962280219851817
Inferring the direction of a causal link and estimating its effect via a Bayesian Mendelian randomization approach
Abstract
The use of genetic variants as instrumental variables - an approach known as Mendelian randomization - is a popular epidemiological method for estimating the causal effect of an exposure (phenotype, biomarker, risk factor) on a disease or health-related outcome from observational data. Instrumental variables must satisfy strong, often untestable assumptions, which means that finding good genetic instruments among a large list of potential candidates is challenging. This difficulty is compounded by the fact that many genetic variants influence more than one phenotype through different causal pathways, a phenomenon called horizontal pleiotropy. This leads to errors not only in estimating the magnitude of the causal effect but also in inferring the direction of the putative causal link. In this paper, we propose a Bayesian approach called BayesMR that is a generalization of the Mendelian randomization technique in which we allow for pleiotropic effects and, crucially, for the possibility of reverse causation. The output of the method is a posterior distribution over the target causal effect, which provides an immediate and easily interpretable measure of the uncertainty in the estimation. More importantly, we use Bayesian model averaging to determine how much more likely the inferred direction is relative to the reverse direction.
Keywords: Bayesian model averaging; Causal inference; Mendelian randomization; genetic epidemiology; instrumental variables; robust estimation; sparsity prior.
Figures
![Figure 1.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig1.gif)
![Figure 2.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig2.gif)
![Figure 3.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig3.gif)
![Figure 4.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig4.gif)
![Figure 5.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig5.gif)
![Figure 6.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig6.gif)
![Figure 7.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig7.gif)
![Figure 8.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig8.gif)
![Figure 9.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig9.gif)
![Figure 10.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig10.gif)
![Figure 11.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig11.gif)
![Figure 12.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig12.gif)
![Figure 13.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig13.gif)
![Figure 14.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig14.gif)
![Figure 15.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig15.gif)
![Figure 16.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig16.gif)
![Figure 17.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig17.gif)
![Figure 18.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig18.gif)
![Figure 19.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig19.gif)
![Figure 20.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig20.gif)
![Figure 21.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig21.gif)
![Figure 22.](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/7221461/bin/10.1177_0962280219851817-fig22.gif)
Similar articles
-
Profile-likelihood Bayesian model averaging for two-sample summary data Mendelian randomization in the presence of horizontal pleiotropy.Stat Med. 2022 Mar 15;41(6):1100-1119. doi: 10.1002/sim.9320. Epub 2022 Jan 20. Stat Med. 2022. PMID: 35060160 Free PMC article.
-
Bayesian variable selection with a pleiotropic loss function in Mendelian randomization.Stat Med. 2021 Oct 15;40(23):5025-5045. doi: 10.1002/sim.9109. Epub 2021 Jun 21. Stat Med. 2021. PMID: 34155684 Free PMC article.
-
[Challenges and factors that influencing causal inference and interpretation, based on Mendelian randomization studies].Zhonghua Liu Xing Bing Xue Za Zhi. 2020 Aug 10;41(8):1231-1236. doi: 10.3760/cma.j.cn112338-20200521-00749. Zhonghua Liu Xing Bing Xue Za Zhi. 2020. PMID: 32867428 Review. Chinese.
-
Extending Causality Tests with Genetic Instruments: An Integration of Mendelian Randomization with the Classical Twin Design.Behav Genet. 2018 Jul;48(4):337-349. doi: 10.1007/s10519-018-9904-4. Epub 2018 Jun 7. Behav Genet. 2018. PMID: 29882082 Free PMC article.
-
A review of instrumental variable estimators for Mendelian randomization.Stat Methods Med Res. 2017 Oct;26(5):2333-2355. doi: 10.1177/0962280215597579. Epub 2015 Aug 17. Stat Methods Med Res. 2017. PMID: 26282889 Free PMC article. Review.
Cited by
-
GWAS advancements to investigate disease associations and biological mechanisms.Clin Transl Discov. 2024 Jul;4(3):e296. doi: 10.1002/ctd2.296. Epub 2024 May 1. Clin Transl Discov. 2024. PMID: 38737752
-
A Bayesian approach to Mendelian randomization using summary statistics in the univariable and multivariable settings with correlated pleiotropy.Am J Hum Genet. 2024 Jan 4;111(1):165-180. doi: 10.1016/j.ajhg.2023.12.002. Am J Hum Genet. 2024. PMID: 38181732 Free PMC article.
-
Mendelian randomization.Nat Rev Methods Primers. 2022 Feb 10;2:6. doi: 10.1038/s43586-021-00092-5. Nat Rev Methods Primers. 2022. PMID: 37325194 Free PMC article.
-
Can metabolomic signature foster precision medicine approaches in nonalcoholic fatty liver disease arena?Hepatobiliary Surg Nutr. 2022 Dec;11(6):886-888. doi: 10.21037/hbsn-22-459. Hepatobiliary Surg Nutr. 2022. PMID: 36523943 Free PMC article. No abstract available.
-
Causal inference on neuroimaging data with Mendelian randomisation.Neuroimage. 2022 Sep;258:119385. doi: 10.1016/j.neuroimage.2022.119385. Epub 2022 Jun 14. Neuroimage. 2022. PMID: 35714886 Free PMC article. Review.
References
-
- Fletcher RH, Fletcher SW, Fletcher GS. Clinical epidemiology: the essentials, Philadelphia: Lippincott Williams & Wilkins, 2012.
-
- Lawlor DA, Harbord RM, Sterne JAC, et al. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Stat Med 2008; 27: 1133–1163. - PubMed
-
- Davey Smith G, Paternoster L, Relton C. When will Mendelian randomization become relevant for clinical practice and public health? JAMA 2017; 317: 589–591. - PubMed
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources