Pleiotropy robust Mendelian randomization: Jack Bowden
The explosion in publicly available data from genome-wide association studies is accelerating the use of Mendelian randomization (MR) in bio-medicine. Summary data estimates of genetic association from large numbers of variants are now being synthesised for causal inference within the two-sample framework. Over the coming years, cohort studies such as UK Biobank will also provide a rich individual-level data resource for MR investigations. In a short time, the field has seen, and will continue to see, a dramatic rise in the power for testing causal hypotheses. There is a justified concern that when large numbers of genetic variants are included in MR analyses, with many lacking a firm biological basis for their association with the exposure, a sizeable proportion of these variants are likely to be invalid instrumental variables.
Aims and Objectives
A chief concern is that genetic variants may exert an effect on the outcome not through the exposure of interest - which is referred to as horizontal pleiotropy. Programme 2 will focus on the development of statistical methods for MR in the presence of pleiotropy and other associated biases. Building upon recent methodological advancements in this field, we will develop analysis tools that: • Visualise MR data to facilitate the detection of pleiotropy; • Explicitly model and adjust for pleiotropy;
- Provide natural robustness to pleiotropy;
- Adjust for non-random selection into (or out of) cohort studies used for MR;
- Novel methods for data visualisation and analysis will be incorporated into statistical software platforms such as MR-Base
Programme 2 links key players in methodological research around the world to leading biomedical scientists at the IEU. It will develop the tools to enable epidemiological researchers to fully capitalise on emerging large-scale data sources whilst preserving the principles and rigour of causal inference.
Research Highlight 1
This was the first paper to propose a formal test for bias due to pleiotropy in MR, and to define a new assumption - the Instrument Strength Independent of Direct Effect (InSIDE) assumption - as weaker alternative to the exclusion restriction, with which to identify causal effects.
Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. International Journal of Epidemiology 2015; 44: 512-525 https://www.ncbi.nlm.nih.gov/pubmed/26050253
Research Highlight 2
This was the first paper to propose a robust summary data MR method that can provide consistent estimates of causal effect when up to 50% of the included genetic variants are invalid instrumental variables.
Bowden J, Davey Smith G, Haycock P, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genetic Epidemiology 2016; 40: 304-314. https://www.ncbi.nlm.nih.gov/pubmed/27061298
Research Highlight 3
This paper describes the statistical theory behind a novel physical interpretation of meta-analysis. The work has since been incorporated into a rudimentary software platform, a public outreach tool and a YouTube video. https://youtu.be/i675gZNe3MY
Bowden J, Jackson C. Weighing evidence `steampunk style’ via the Meta-Analyser. The American Statistician 2016; 70: 385-394 goo.gl/8Zxuzy
Research Highlight 4
This paper generalises two-stage uniform minimum variance unbiased estimation for multivariate normal outcomes, enabling simultaneous adjustment for bias due to selection and correlation. Applications include multi-arm, multi-stage adaptive trials and genome-wide association studies.
Robertson D, Prevost T, Bowden J. Accounting for selection and correlation in the analysis of two-stage genome-wide association studies. Biostatistics 2016; 17: 634-49 https://www.ncbi.nlm.nih.gov/pubmed/26993061
Research Highlight 5
This paper proposed a more efficient version of the log-rank test in clinical trials assessing a time-to-event outcome in the presence of non-compliance. It can be used as a tool for quantifying the health economic benefit of treatments.
Bowden J, Seaman S, Huang X, White I. Gaining power and precision by using model-based weights in the analysis of late stage cancer trials with substantial treatment switching. Statistics in Medicine 2016; 35: 1423-1440. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4871231/