Age-varying genetic associations and implications for bias in Mendelian randomization analyses
Labrecque, J. A.; Swanson, S. A.
Show abstract
Estimates from conventional Mendelian randomization (MR) analyses can be biased when the genetic variants proposed as instruments vary over age in their relationship with the exposure. For four exposures commonly studied using MR, we assessed the degree to which their relationship with genetic variants commonly used as instruments varies by age using flexible, spline-based models in UK Biobank data. Using these models, we then estimated how biased MR estimates would be due to age-varying relationships using plasmode simulations. We found that most genetic variants had age-varying relationships with the exposure for which they are a proposed instrument. Body mass index and LDL cholesterol had the most variation while alcohol consumption had very little. This variation over age led to small potential biases in some cases (e.g. alcohol consumption and C-reactive protein) and large potential biases for many proposed instruments for BMI and LDL.
Matching journals
The top 2 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Bias in two-sample Mendelian randomization when using heritable covariable-adjusted summary associations 95%
- The use of negative control outcomes in Mendelian Randomisation to detect potential population stratification or selection bias. 94%
- Estimation of time-varying causal effects with multivariable Mendelian randomization: some cautionary notes 94%
Similar papers in this journal
- Mendelian randomisation for mediation analysis: current methods and challenges for implementation 95%
- Meta-regression of Genome-Wide Association Studies to estimate age-varying genetic effects 93%
- Exploring and mitigating potential bias when genetic instrumental variables are associated with multiple non-exposure traits in Mendelian randomization 92%
Similar papers in this journal
- Analyses using multiple imputation need to consider missing data in auxiliary variables 93%
- A structural mean modelling Mendelian randomization approach to investigate the lifecourse effect of adiposity: applied and methodological considerations 93%
- Lifecourse Socioeconomic Status and Cognitive Performance Among Midlife Latina Women in a California Agricultural Region 90%
Similar papers in this journal
- Relaxing parametric assumptions for non-linear Mendelian randomization using a doubly-ranked stratification method 96%
- Estimation of causal effects of a time-varying exposure at multiple time points through Multivariable Mendelian randomization 95%
- Assortative mating and within-spouse pair comparisons 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.