Mendelian randomization accounting for correlated and uncorrelated pleiotropic effects using genome-wide summary statistics

Morrison, Jean; Knoblauch, Nicholas W.; Marcus, Joseph; Stephens, Matthew; He, Xin

doi:10.1038/s41588-020-0631-4

Cited by 441 publications

(576 citation statements)

References 32 publications

Supporting

Mentioning

562

Contrasting

Order By: Relevance

“…In terms of results, in simulations we have seen that our approach remains robust (i.e. unbiased) to increased polygenicity (up to 100%) of the confounder U , while CAUSE has substantial false positive rate already when ⇡ u > 0.3 in samples of size 40,000 (see Fig4a of Morrison et al [10] ).…”

Section: Application To Association Summary Statistics Of Complex Traitsmentioning

confidence: 74%

“…However, such instruments will have a direct e↵ect on the outcome that is correlated to their instrument strength, violating the InSIDE assumption. A very recent approach (CAUSE) proposes a structural equation mixed e↵ect model [10] , similar to ours. In the Discussion, we will provide a detailed account of the di↵erences between CAUSE and our method.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Simultaneous estimation of bi-directional causal effects and heritable confounding from GWAS summary statistics

Darrous

Mounier

Kutalik

2020

Preprint

View full text Add to dashboard Cite

Mendelian Randomisation (MR), an increasingly popular method that estimates the causal effects of risk factors on complex human traits, has seen several extensions that relax its basic assumptions. However, most of these extensions suffer from two major limitations; their under-exploitation of genome-wide markers, and sensitivity to the presence of a heritable confounder of the exposure-outcome relationship. To overcome these limitations, we propose a Latent Heritable Confounder MR (LHC-MR) method applicable to association summary statistics, which estimates bi-directional causal effects, direct heritability, and confounder effects while accounting for sample overlap. We demonstrate that LHC-MR outperforms several existing MR methods in a wide range of simulation settings and apply it to summary statistics of 13 complex traits. Besides several concordant results, LHC-MR unravelled new mechanisms (how being diagnosed for certain diseases might lead to improved lifestyle) and revealed potential false positive findings of standard MR methods (apparent causal effect of body mass index on educational attainment may be driven by a strong ignored confounder). Phenome-wide search to identify LHC-implied heritable confounders showed remarkable agreement between the LHC-estimated causal effects of the latent confounder and those for the potentially identified ones. Finally, LHC-MR naturally decomposes genetic correlation to causal effect-driven and confounder-driven contributions, demonstrating that the genetic correlation between systolic blood pressure and diabetes is predominantly confounder-driven.

show abstract

Section: Application To Association Summary Statistics Of Complex Traitsmentioning

confidence: 74%

Section: Introductionmentioning

confidence: 99%

Simultaneous estimation of bi-directional causal effects and heritable confounding from GWAS summary statistics

Darrous

Mounier

Kutalik

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…A Mendelian randomization (MR) approach is one possible avenue to address confounding issues. Lipids and CHD have been investigated using MR, and the results suggest a causal association between LDL and increased CHD but a null association for HDL . Less investigated is the association between lipids and other competing risks, and more broadly, survival to old age.…”

Section: Discussionmentioning

confidence: 99%

Associations between Serum Levels of Cholesterol and Survival to Age 90 in Postmenopausal Women

Maihofer

Shadyab

Wild

et al. 2020

J American Geriatrics Society

View full text Add to dashboard Cite

OBJECTIVES: Although elevated lipid levels predict increased risk of coronary heart disease and death in middle-aged women and men, evidence is mixed if lipid levels measured in later life predict survival to very old ages. We examined lipid levels and survival to age 90 with or without intact mobility in a large cohort of older women. DESIGN: Prospective cohort. SETTING: Laboratory collection at a Women's Health Initiative (WHI) center and longitudinal follow-up via mail. PARTICIPANTS: Women aged 68 to 81 years at baseline. MEASUREMENTS: Serum high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol were collected at baseline. Participant survival status and selfreported mobility was compared across lipid levels. RESULTS: HDL and LDL levels were not associated with survival to age 90 after adjustment for cardiovascular risk factors (HDL: quartile (Q) 2: odds ratio [OR] = 1.14 [95% confidence interval [CI] = .94-1.38]; Q3 OR = 1.08 [95% CI = .88-1.33]; Q4 OR = 1.09 [95% CI = .88-1.35]; LDL: Q2 OR = 1.07 [95% CI = .88-1.31]; Q3 OR = 1.27 [95% CI = 1.04-1.55]; Q4 OR = 1.07 [95% CI = .88-1.31]). Similarly, no associations were observed between HDL and LDL levels and survival to age 90 with mobility disability. High HDL was not associated with survival to age 90 with intact mobility after adjustment for other cardiovascular risk factors. Compared with the lowest LDL quartile, the three upper LDL quartiles were associated with greater odds of survival to age 90 with intact mobility (LDL: Q2 OR = 1.31 [95% CI = .99-1.74]; Q3 OR = 1.43 [95% CI = 1.07-1.92]; Q4 OR = 1.35 [95% CI = 1.01-1.80]; P = .05).CONCLUSION: Neither higher HDL nor lower LDL levels predicted survival to age 90, but higher LDL predicted healthy survival. These findings suggest the need for reevaluation of healthy LDL levels in older women. J Am Geriatr Soc 68:288-296, 2020.

show abstract

“…Finally unlike both, we are able to estimate both effect directions simultaneously, allowing our model to accommodate graphs with cycles. Our method is also related to the recently proposed CAUSE [55]. There, the authors model observed SNPs effects as coming from a mixture of pleiotropic and non-pleiotropic SNPs.…”

Section: Discussionmentioning

confidence: 99%

Phenome-scale causal network discovery with bidirectional mediated Mendelian randomization

Brown

Knowles

2020

Preprint

View full text Add to dashboard Cite

Inference of directed biological networks from observational genomics datasets is a crucial but notoriously difficult challenge. Modern population-scale biobanks, containing simultaneous measurements of traits, biomarkers, and genetic variation, offer an unprecedented opportunity to study biological networks. Mendelian randomization (MR) has received attention as a class of methods for inferring causal effects in observational data that uses genetic variants as instrumental variables, but MR methods rely on assumptions that limit their application to complex traits at the biobank-scale. Moreover, MR estimates the total effect of one trait on another, which may be mediated by other factors. Biobanks include measurements of many potential mediators, in principle enabling the conversion of MR estimates into direct effects representing a causal network. Here, we show that this can be accomplished by a flexible two stage procedure we call bidirectional mediated Mendelian randomization (bimmer). First, bimmer estimates the effect of every trait on every other. Next, bimmer finds a parsimonious network that explains these effects using direct and mediated causal paths. We introduce novel methods for both steps and show via extensive simulations that bimmer is able to learn causal network structures even in the presence of non-causal genetic correlation. We apply bimmer to 405 phenotypes from the UK biobank and demonstrate that learning the network structure is invaluable for interpreting the results of phenome-wide MR, while lending causal support to several recent observational studies. * bb2991@columbia.edu † dak2173@columbia.edu causal effects, MR methods must make strong assumptions that limit their ability to be applied at the biobank-scale. Perhaps the most controversial assumption is that the SNP only effects B through A (i.e. there is no horizontal pleiotropy). Recent methods such as Egger regression and the mode-based-estimator are able to relax this assumption, instead assuming there is no correlated pleiotropy or modal pleiotopy, respectively [11,12]. Another approach, the latent causal variable (LCV) model, is able to detect causality under arbitrarily-structured pleiotropy [13]. However, the quantity that LCV calculates is not interpretable as the causal effect size of A on B. Most MR studies also presuppose the direction of effect, specifying one phenotype as the outcome and the other as the exposure. Pre-specifying the effect direction is sound when the outcome is clearly biologically downstream of the exposure, but in some cases it is better to learn the direction of the effect from the data. Some researchers have instead used bidirectional MR [14,15], which tests for an effect in each direction, or gwas-pw [16], which infers the effect direction from the data. However, the utility of these approaches for complex traits, which might contain non-causal genetic correlation, is questionable [13].In mimicking a randomized controlled trial, MR estimates the total causal effect (TCE) of A on B [17]. This effec...

show abstract

Mendelian randomization accounting for correlated and uncorrelated pleiotropic effects using genome-wide summary statistics

Cited by 441 publications

References 32 publications

Simultaneous estimation of bi-directional causal effects and heritable confounding from GWAS summary statistics

Simultaneous estimation of bi-directional causal effects and heritable confounding from GWAS summary statistics

Associations between Serum Levels of Cholesterol and Survival to Age 90 in Postmenopausal Women

Phenome-scale causal network discovery with bidirectional mediated Mendelian randomization

Contact Info

Product

Resources

About