Skip to contents

Summary statistics from all five test cohorts with annotations

Source: R/map.R
benchmark_results.Rd

This table summary statistics for PGRM association in BioVU (EUR and AFR), BBJ, MGI, and UKBB. All associations are annotated and the table is in long format.

Usage

benchmark_results

Format

A data.table with the following columns:

  • assoc_ID: Integer vector of the unique identifier for each association

  • phecode: Character vector of the phecode for the association

  • phecode_string: Character vector of the phecode string label

  • category_string: Character vector of the phecode category

  • cases: Integer of the number of cases for specified phecode/cohort

  • controls: Integer of the number of controls for specified phecode/cohort

  • odds_ratio: Numeric odds ratio for test cohort association

  • P: Numeric raw P-value for test cohort association

  • L95: Numeric 95% lower confidence interval for test cohort association

  • U95: Numeric 95% upper confidence interval for test cohort association

  • cohort: Character vector of the source cohort (BBJ: Biobank Japan, BioVU_AFR: BioVU, African ancestry, BioVU_EUR: BioVU, European ancestry, MGI: Michigan genomics initiative, UKBB: UK Biobank)

  • ancestry: Character vector of the ancestery of the source GWAS using 1000 Genomes superpopulations ancestry groupings (AFR: African, EAS: East Asian, EUR: European, AMR: AdMixed American, SAS: South Asian)

  • rsID: Character indicating the Reference SNP cluster ID

  • risk_allele_dir: Character indicating the specifying the risk allele direction (alt: Alternate allele, ref: Reference allele)

  • RAF: GnomAD frequency of risk allele, matched to ancestry

  • cat_LOG10_P: Numeric vector of the -log10(P) of the association from the catalog

  • cat_OR: Numeric vector of the odds ratio of the association

  • cat_L95: Numeric vector of the 95% lower confidence interval of the association

  • cat_U95: Numeric vector of the 95% upper confidence interval of the association

  • Study_accession: Character vector of the study accession ID from the GWAS catalog

  • Powered: Boolean value indicating if the association is powered at >=80%

  • rep: Boolean value indicating if the association replicated

  • CI_overlap: Character vector indicating the overlap of the confidence intervals from the catalog and test association

  • CI_overlap: Character vector indicating the test cohort

  • pub_count: The number of times the association has been reported in the catalog

  • first_pub_date: The first date the association was published