Print  Close

Mendelian randomization using summary statistics from two homogeneous or heterogeneous samples

Abstract Number:

2012 

Submission Type:

Contributed Abstract 

Contributed Abstract Type:

Paper 

Participants:

Kai Wang (1), Grace Wang (2)

Institutions:

(1) University of Iowa, N/A, (2) Department of Biostatistics, Harvard T.H. Chan School of Public Health, United States

Co-Author:

Grace Wang  
Department of Biostatistics, Harvard T.H. Chan School of Public Health

First Author:

Kai Wang  
University of Iowa

Presenting Author:

Kai Wang  
University of Iowa

Abstract Text:

Despite its increasing popularity, some important theoretical properties of Mendelian randomization (MR) remain unclear, even when using summary statistics from two homogeneous samples. We first derive an explicit expression for the two-stage least squares (TSLS) estimator and its asymptotic variance in terms of summary statistics from an exposure GWAS and an outcome GWAS, without making the approximations that the inverse variance weighted (IVW) method and its various variations make. When the instrument SNPs are mutually in linkage equilibrium, this estimator is a weighted average of the Wald ratios, with the weights proportional to the estimated SNP heritability for the exposure. Therefore, contrary to a long-standing misconception, the TSLS estimator is not asymptotically equivalent to the IVW estimator. We further generalize this TSLS estimator to the case where the two GWAS samples are heterogeneous. Simulation studies are used to validate our results, and data applications are employed to illustrate the proposed method.

Keywords:

Mendelian randomization
|summary statistics
|two-stage least-squares regression
|inverse variance weighting
|generalized method of moment|

Sponsors:

Section on Statistics in Genomics and Genetics

Tracks:

Miscellaneous

Can this be considered for alternate subtype?

Yes

Are you interested in volunteering to serve as a session chair?

Yes

I have read and understand that JSM participants must abide by the Participant Guidelines.

Yes

I understand that JSM participants must register and pay the appropriate registration fee by June 3, 2025. The registration fee is non-refundable.

I understand