In mglev1n/locusplotr: Create Regional Association Plots

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

locusplotr

The goal of locusplotr is to allow users to integrate genome-wide association study results with linkage disequilibrium data to create regional association plots surrounding a genomic locus of interest. The gg_locusplot function allows the user to provide GWAS summary statistics for a region of interest and a reference/sentinel/lead variant within that region, and will return a ggplot object with a regional association plot. In the background, this package uses a helper function named ld_extract_locuszoom to query the University of Michigan LocusZoom API (https://portaldev.sph.umich.edu/) to obtain linkage disequilibrium data for a genetic variant and genomic region of interest (allowing the user to specify genome build and genetic ancestry).

Installation

You can install locusplotr from GitHub:

# install.packages("devtools")
devtools::install_github("mglev1n/locusplotr")

Usage

Start with a dataframe containing genome wide association study summary statistics at a locus of interest

library(locusplotr)

fto_locus_df

Use the gg_locusplot function to create a regional association plot. This calls the ld_extract_locuszoom function in the background to fetch linkage disequilibrium data from the University of Michigan LocusZoom API (https://portaldev.sph.umich.edu/).

library(tidyverse)

fto_locus_df %>%
  gg_locusplot(
    lead_snp = "rs62033413",
    rsid = rsid,
    chrom = chromosome,
    pos = position,
    ref = effect_allele,
    alt = other_allele,
    p_value = p_value
  )

Use the plot_genes argument to include a plot of genes within the region beneath the main regional association plot.

gg_locusplot(
  df = fto_locus_df,
  lead_snp = "rs62033413",
  rsid = rsid,
  chrom = chromosome,
  pos = position,
  ref = effect_allele,
  alt = other_allele,
  p_value = p_value,
  plot_genes = TRUE
)

Use the plot_recombination argument to include a secondary axis containing recombination rates on regional association plot.

gg_locusplot(
  df = fto_locus_df,
  lead_snp = "rs62033413",
  rsid = rsid,
  chrom = chromosome,
  pos = position,
  ref = effect_allele,
  alt = other_allele,
  p_value = p_value,
  plot_genes = TRUE,
  plot_recombination = TRUE
)

Additional arguments allow the user to specify the genome build (GRCh37, GRCH38), population (Eg. ALL, AMR, AFR, EUR, EAS, etc.), plot title/subtitle, and directory to save the regional association plot as a .pdf.

The ld_extract_locuszoom function, called in the background by gg_locusplot, can also be used to retrieve linkage disequilibrium data for a genetic variant and genomic region of interest. The user may again specify genome build, population, and the specific linkage disequilibrium metric (Eg. r, rsquare, cov)

ld_extract_locuszoom(chrom = 16, pos = 53830055, ref = "C", alt = "G", start = 53830055 - 5e5, stop = 53830055 + 5e5, genome_build = "GRCh37", population = "ALL", metric = "rsquare")

mglev1n/locusplotr documentation built on July 31, 2024, 8:41 p.m.