compare_effect_to_gwascatalog2: Compare the genetic effect sizes in the test dataset to the...
In MRCIEU/mrQC: CheckSumStats

compare_effect_to_gwascatalog2

R Documentation

Compare the genetic effect sizes in the test dataset to the GWAS catalog

Description

Compare the direction of effects and effect allele frequency between the test dataset and the GWAS catalog, in order to identify effect allele meta data errors

Usage

compare_effect_to_gwascatalog2(
  dat = NULL,
  efo = NULL,
  efo_id = NULL,
  trait = NULL,
  gwas_catalog_ancestral_group = c("European", "East Asian"),
  exclude_palindromic_snps = TRUE,
  map_association_to_study = FALSE,
  beta = "beta",
  se = "se",
  gwas_catalog = NULL,
  force_all_trait_study_hits = FALSE,
  distance_threshold = distance_threshold
)

Arguments

`dat`	the test dataset of interest
`efo`	trait of interest in the experimental factor ontology
`efo_id`	ID for trait of interest in the experimental factor ontology
`trait`	the trait of interest
`gwas_catalog_ancestral_group`	restrict the comparison to these ancestral groups in the GWAS catalog. Default is set to (c("European","East Asian")
`exclude_palindromic_snps`	should the function exclude palindromic SNPs? default set to TRUE. If set to FALSE, then conflicts with the GWAS catalog could reflect comparison of different reference strands.
`map_association_to_study`	map associations to study in GWAS catalog. This supports matching of results on PMID and study ancestry, which increases accuracy of comparisons, but is slow when there are large numbers of associations. Default = FALSE.
`beta`	name of the column containing the SNP effect size
`se`	name of the column containing the standard error for the SNP effect size.
`gwas_catalog`	user supplied data frame containing results from the GWAS catalog for the trait of interest. If set to NULL then the function will retrieve results from the GWAS catalog.
`force_all_trait_study_hits`	force the comparison to include GWAS hits from the test dataset if they are not in the GWAS catalog? This should be set to TRUE only if dat is restricted to GWAS hits for the trait of interest. This is useful for visualising whether the test trait study has an unusually larger number of GWAS hits, which could, in turn, indicate analytical issues with the summary statistics
`distance_threshold`	distance threshold for deciding if the GWAS hit in the test dataset is present in the GWAS catalog. For example, a distance_threshold of 25000 means that the GWAS hit in the test dataset must be within 25000 base pairs of a GWAS catalog association, otherwise it is reported as missing from the GWAS catalog.