Getting the most out of MR-Base: a platform for Mendelian randomization using results from genome-wide association studies
MR-Base is a platform for two-sample Mendelian randomization (MR) using results from genome-wide association studies (MR Base). It includes a large database of curated summary association statistics from >1000 GWAS analyses combined with an automated analysis pipeline (implemented in R, including web and command line interfaces). This workshop will teach users how to get the most out of MR-Base in their research.
Tutors
The tutors on the workshop are based at the MRC Integrative Epidemiology Unit and are the lead developers of the MR-Base platform. They include:
- Philip Haycock (CRUK Population Research Fellow, MRC IEU Research Fellow)
- Gibran Hemani (Wellcome Senior Research Fellow)
- Jie Zheng (Vice-Chancellor’s Fellow)
- Tom Gaunt (Professor in Health & Biomedical Informatics)
Workshop assistants
- Denis Baird (Senior Research Associate)
- Valeriia Haberland (Senior Research Associate)
Venue
School of Chemistry, Cantocks Close, Bristol, BS8 1TS
Date and times
The workshop will take place on Friday 19th July 2019, from 13:30 to 17:00.
Workshop fee
The fee for attending the workshop is £20.
Aims and objectives
The aim of this workshop is to show users how to get the most out of MR-Base for their research. Users will learn about:
- recommended workflows
- how to use the MR-Base website and two sample MR R package
- the different types of analyses that can be implemented, including focused MR, phenome-wide MR on drug targets and multivariable MR
- interpretation of outputs, including consideration of sensitivity analyses and potential violations of MR assumptions, including MR-Egger, weighted median, mode estimate and Steiger filtering approach
- Interaction section for possible open questions, including new methods development, public health and MR in drug development
Workshop outline
The workshop will last three and a half hours. The first half will be spent covering the functionality of MR-Base, including a MR study of lipids and human health as an applied example. In the second half users will be given the opportunity to implement their own analyses of interest. Experienced users and developers of MR-Base will be on hand to give detailed support and feedback. The course will not cover the theoretical underpinnings or limitations of MR in detail.
By the end of the course students will be able to:
- use MR-Base to perform focused and hypothesis-free two-sample MR analyses
- interpret the robustness of their results to violations of MR assumptions.
Who the course is intended for?
The course is primarily intended for users with some experience of MR and who wish to maximise the efficiency of their MR workflows. It is assumed that users already have some understanding of the theoretical underpinnings, assumptions and limitations of MR. The workshop would ideally suit researchers who have attended the MR short course on Tuesday, or with equivalent experience.
*It is highly recommended that delegates bring their own laptops to the workshop in order to follow along with the presentation and practice using the application with their data.