Skip to contents

Improved meta-analysis using tidyGWAS:ed files

meta_analyse.Rd

meta_analyse() will:

  • Flip the effect allele to be the reference allele (using either GRCh37 or GRCh38), control with ref

  • variant id is constructed using RSID:EffectAllele:OtherAllele (Which is now RSID:REF:ALT)

  • EAF (allele frequency) and INFO (imputation quality) are weighted by sample size, if present

  • CaseN, N, ControlN and EffectiveN are all summed and carried forward

Usage

meta_analyse(
  ds,
  chromosomes = c(1:22),
  min_EAF = NULL,
  ref = c("REF_38", "REF_37"),
  safe_mode = FALSE
)

Arguments

ds

a dataset object, see arrow::open_dataset()

chromosomes

Which chrosomes to apply meta-analysis across. Default is autosomes: 1:22

min_EAF

Filter on minimal EAF (effect allele frequency) value. (0 - 0.5)

ref

Reference genome version to use for reference allele

safe_mode

Apply additional filters to ensure no missing/inf/NaN across key columns? This is set to FALSE by default because these filters should already be guaranteed with the use of tidyGWAS()

Value

a dplyr::tibble()

Examples

if (FALSE) { # \dontrun{
meta_analyse2(ds, min_EAF = 0.01)
} # }