• Corpus ID: 88521372

Mendelian Randomization when Many Instruments are Invalid: Hierarchical Empirical Bayes Estimation

@article{Li2017MendelianRW,
  title={Mendelian Randomization when Many Instruments are Invalid: Hierarchical Empirical Bayes Estimation},
  author={Sai Li},
  journal={arXiv: Methodology},
  year={2017}
}
  • Sai Li
  • Published 5 June 2017
  • Computer Science
  • arXiv: Methodology
Estimating the causal effect of an exposure on an outcome is an important task in many economical and biological studies. Mendelian randomization, in particular, uses genetic variants as instruments to estimate causal effects in epidemiological studies. However, conventional instrumental variable methods rely on some untestable assumptions, which may be violated in real problems. In this paper, we adopt a Bayesian framework and build hierarchical models to incorporate invalid effects of… 

Figures from this paper

INFERENCE IN TWO-SAMPLE SUMMARY-DATA MENDELIAN RANDOMIZATION USING ROBUST ADJUSTED PROFILE SCORE By
TLDR
This paper studies statistical inference in the increasingly popular two-sample summarydata MR design and proposes a consistent and asymptotically normal estimator by adjusting the profile score, and tackles the idiosyncratic pleiotropy by robustifying the adjusted profile score.
Inferring the direction of a causal link and estimating its effect via a Bayesian Mendelian randomization approach
TLDR
A Bayesian approach called BayesMR is proposed that is a generalization of the Mendelian randomization technique in which it allows for pleiotropic effects and, crucially, for the possibility of reverse causation.
Statistical inference in two-sample summary-data Mendelian randomization using robust adjusted profile score
TLDR
This paper studies statistical inference in the increasingly popular two-sample summary-data Mendelian randomization, finding strong evidence of both systematic and idiosyncratic pleiotropy in MR, echoing some recent discoveries in statistical genetics.
Causal Inference for Nonlinear Outcome Models with Possibly Invalid Instrumental Variables
Instrumental variable methods are widely used for inferring the causal effect of an exposure on an outcome when the observed relationship is potentially affected by unmeasured confounders. Existing
Testing and controlling for horizontal pleiotropy with probabilistic Mendelian randomization in transcriptome-wide association studies
TLDR
A powerful TWAS method based on probabilistic Mendelian Randomization, PMR-Egger, which is reasonably robust under various types of model misspecifications, is more powerful than existing TWAS/MR approaches, and can directly test for horizontal pleiotropy.
Inferring Causal Relationships Between Risk Factors and Outcomes from Genome-Wide Association Study Data
TLDR
This work reviews the Mendelian randomization paradigm for making causal inferences using genetic variants, and considers monogenic analysis, in which genetic variants are taken from a single gene region, and polygenic analysis, which includes variants from multiple regions.
Inferring causal relationships between risk factors and outcomes from GWAS data
TLDR
The Mendelian randomization paradigm for making causal inferences using genetic variants is introduced and monogenic analysis, in which genetic variants are taken from a single gene region, and polygenic analysis, which include variants from multiple regions are considered.
Data integration in causal inference
TLDR
Recent advances on combining randomized clinical trial with external information from observational studies or historical controls, combining samples when no single sample has all relevant variables with application to two-sample Mendelian randomization are summarized.
Testing and controlling for horizontal pleiotropy with the probabilistic Mendelian randomization in transcriptome-wide association studies
TLDR
PMR-Egger is presented, which relies on an MR likelihood framework that unifies many existing TWAS and MR methods, accommodates multiple correlated instruments, tests the causal effect of gene on trait in the presence of horizontal pleiotropic effects, and is scalable to hundreds of thousands of individuals.
Transcriptome-wide association studies: a view from Mendelian randomization
TLDR
A technical review on thirteen TWAS methods and shows that these methods can all be viewed as two-sample Mendelian randomization (MR) analysis, which has been widely applied in GWASs for examining the causal effects of exposure on outcome.
...
...

References

SHOWING 1-10 OF 43 REFERENCES
Instrumental Variables Estimation With Some Invalid Instruments and its Application to Mendelian Randomization
TLDR
A fast penalized ℓ1 estimation method, called sisVIVE, is introduced for estimating the causal effect without knowing which instruments are valid, with theoretical guarantees on its performance.
Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator
TLDR
A novel weighted median estimator for combining data on multiple genetic variants into a single causal estimate is presented, which is consistent even when up to 50% of the information comes from invalid instrumental variables.
Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology
TLDR
The use of germline genetic variants that proxy for environmentally modifiable exposures as instruments for these exposures is one form of IV analysis that can be implemented within observational epidemiological studies and can be considered as analogous to randomized controlled trials.
Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression
TLDR
An adaption of Egger regression can detect some violations of the standard instrumental variable assumptions, and provide an effect estimate which is not subject to these violations, and provides a sensitivity analysis for the robustness of the findings from a Mendelian randomization investigation.
Mendelian randomization: genetic anchors for causal inference in epidemiological studies
TLDR
Developments of MR, including two-sample MR, bidirectional MR, network MR, two-step MR, factorial MR and multiphenotype MR, are outlined in this review.
Methodological Challenges in Mendelian Randomization
TLDR
Critical attention is given to the assumptions underlying Mendelian randomization analysis and their biological plausibility and to the use of sensitivity analysis in evaluating the consequences of violations in the assumptions and in attempting to correct for those violations.
Network Mendelian randomization: using genetic variants as instrumental variables to investigate mediation in causal pathways
TLDR
These Mendelian randomization methods can be used to estimate direct and indirect causal effects in a mediation setting, and have potential for the investigation of more complex networks between multiple interrelated exposures and disease outcomes.
A framework for the investigation of pleiotropy in two‐sample summary data Mendelian randomization
TLDR
How established methods of meta‐regression and random effects modelling from mainstream meta‐analysis are being adapted to perform MR analyses are clarified, and the ability of two popular random effects models to provide robustness to pleiotropy under the IVW approach is investigated.
Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants.
TLDR
The feasibility of well-powered, unbiased MR studies will depend upon the amount of variance in the exposure that can be explained by known genetic factors and the 'strength' of the IV set derived from these genetic factors.
Pleiotropy-robust Mendelian randomization.
TLDR
Pleiotropy-robust Mendelian randomization (PRMR) allows for instrumental variables that violate the exclusion restriction due to pleiotropic working, and it corrects for pleiotropy in the estimation of the causal effect.
...
...