Please specify metabolites (exposure) and outcome of interest

Causal analysis is based on Mendelian randomization (MR) which leverages genetic variants such as single-nucleotide polymorphisms (SNPs) as instrumental variables (IV) to estimate exposure-outcome associations. The growing number of mGWAS studies and two-sample MR method permit causal analysis between metabolite and outcome of interest as described below:

1. Identify SNPs that are significantly associate with a metabolite of interest from our large collections of the recent mGWAS studies (covering > 4000 metabolites including their ratios);
2. Obtain the estimates of associations between these same SNPs with an outcome of interest from public repository. We use Open GWAS Project.
3. Perform SNP filtering and harmonize the effect sizes for SNPs on the exposures and the outcomes to be for the same reference allele.
4. Conduct MR analysis, sensitivity analyses, and explore the graphical outputs

1. Select a metabolite of interest (exposure):

Due to its complex and computing intensive nature, MR analysis is typically performed with one metabolite/exposure at a time, to make sure that each step is performed properly as well as to avoid performance issue (max 5). To increase the chance of identification of the genetically influenced metabolites, metabolites that have passed the threshold in any of the mGWAS studies will be listed below. You will be able to examine the detailed SNP-metabolite associations (studies, p-values, etc) in the next page to fine control which ones to be included for MR analysis.