This function is a wrapper around MR-MEGA, which uses meta-regression to combine GWAS summart statistics while accounting for study/population-specific components of genetic variation. The method was described in Magi et al. (Human Molecular Genetics 2017; https://doi.org/10.1093/hmg/ddx280), and the package can be obtained from the Estonian Genome Centre (https://genomics.ut.ee/en/tools).
Usage
mr_mega(
sumstats_files,
mr_mega_bin,
marker_col = MARKER,
chr_col = CHROM,
pos_col = POS,
effect_allele_col = EFFECT_ALLELE,
other_allele_col = OTHER_ALLELE,
eaf_col = EAF,
beta_col = BETA,
se_col = SE,
p_value_col = P,
samplesize_col = N,
n_pcs = 3
)Arguments
- sumstats_files
List of files containing GWAS summary statistics. These files should all contain the same column header.
- mr_mega_bin
Path to MR-MEGA binary
- marker_col
Column containing unique markers for each variant
- chr_col
Column containing the chromosome
- pos_col
Column containing the position
- effect_allele_col
Column containing the effect allele
- other_allele_col
Column containing the non-effect allele
- eaf_col
Column containing effect allele frequencies
- beta_col
Column containing effect estimates
- se_col
Column containing standard errors
- p_value_col
Column containing p-value
- samplesize_col
Column containing sample size
- n_pcs
Number of genetic principal components to include in the meta-regression.
See also
Other GWAS meta-analysis:
metal_config(),
metal_run()
Examples
if (FALSE) {
mr_mega(sumstats_files = list("/path/to/sumstats_1.txt.gz", "/path/to/sumstats_2.txt.gz"), mr_mega_bin = "/path/to/MR-MEGA")
}