MR uncoupling scripts and updated results from Martin et al. (2022)

Scripts used to conduct analyses for Martin et al. (2022)¹ and updated results based on larger GWAS releases since publication.

R scripts used to conduct analyses and generate results tables

Scripts are split into three parts:

1-MR_analysis.R

Conducts Mendelian Randomisation for each disease and sub-disease using IVW, weighted median, MR-Egger and penalised weighted median methods. Generates Supplementary File 1g(i-iv). Requires the following arguments and will need to be run per exposure trait (BMI, BFP, FA or UFA):

Rscript 1-MR_analysis.R test_trait snp_info summ_stats_file

where test_trait is the name of the exposure (BMI, BFP, FA or UFA),

snp_info is the location and name of the exposure summary statistics file, and

summ_stats_file is the location and name of the file containing individual BetaYG and seBetaYG for the disease outcomes.

N.B. The snp_info file should contain the summary statistics of the exposure genetic variants (as given in Supplementary File 1d) as well as appropriate proxy variants, and the summ_stats_file file should contain the exposure genetic variants (or best available proxies) extracted from the disease outcome GWAS summary statistics.

2-meta_analysis.R

Conducts random-effects meta-analysis of published GWAS and FinnGen IVW MR results together for appropriate matching outcomes. Generates Supplementary File 1e, 1f and 1h. Requires the output from 1-MR_analysis.R as well as a file of diseases to be meta-analysed together (including meta-analysed trait name), titled disease_meta_analysis.txt, of the form:

Trait     Meta_name
CAD_GWAS  CAD_meta
CAD_FG    CAD_meta
PCOS_GWAS PCOS_meta
...       ...

3-correction_FDR_BH.R

Calculates Benjamini-Hochberg false discovery rates of the 37 disease IVW MR results. Requires output from 2-meta_analysis.R only.

Updated results based on larger GWAS releases since publication

The figures and supplementary files corresponding to updated results based on larger FinnGen and published GWAS releases since publication can be found in the updated_results_* folders. The write-ups corresponding to these releases can be found in the Comments section of the publication (accessible here: https://elifesciences.org/articles/72452). The FinnGen releases used for each update are given below:

updated_results_october2022 (to be uploaded soon) - FinnGen release R7.

updated_results_april2022 - FinnGen release R6.

Footnotes

1. Susan Martin, Jessica Tyrrell, Elizabeth Thomas, Matthew Bown, Andrew Wood, Robin Beaumont, Lam Tsoi, Philip E Stuart, James Elder, et al. Disease consequences of higher adiposity uncoupled from its adverse metabolic effects using mendelian randomisation. eLife, 11:e72452, 2022 (https://elifesciences.org/articles/72452). ↩