Generate regional views of GWAS results from whole genome sequence data.
Maintainer: Tim Majarian
Version: 0.1
This package provides functions for plotting summary statistics from GWAS with a particular focus on whole genome sequence and functional data. P-values, genomic context, and multiple bed file tracks can be plotted over small or large regions of the genome. LD information can also be displayed.
Summary statistics
The minimum data required to generate a plot are per-variant GWAS summary statistics. Four values must be defined for each variant:
- Variant id: a unique identifier for each variant
- Chromosome: the chromosome location of each variant
- Position: the genomic position of each variant
- P-value: a p-value for each variant (or some other data to plot on the Y-axis)
| MarkerName | chr | pos | pvalue |
|---|---|---|---|
| rs12345 | 1 | 234030 | 0.803 |
| rs54321 | 10 | 1268346 | 0.001 |
| rs98765 | 22 | 68263057 | 0.283 |
LD
LD information can also be incorperated into the plot. To do so, a file with LD-values per variant needs to be input. This package is designed to be used in conjuction with and handles outputs from our LDGds pipeline. LD data should be stored in a plain, delimited text file as either a matrix or a row vector. Either format must have variant identifiers as row and column names. In conjunction with the LD data, a reference variant must be specified. See the package vignette and accompanying data for an example of input LD data.
| rs12345 | rs54321 | rs98765 | |
|---|---|---|---|
| rs12345 | 1 | 0.0001 | 0.1001 |
| rs54321 | 0.0001 | 1 | 0.82 |
| rs98765 | 0.1001 | 0.82 | 1 |
To install this package, ensure that devtools is installed. Then run:
devtools::install_github("manning-lab/WGSregionalPlot")
library(WGSregionalPlot)Note: See further documentation for advanced options.
Generates a regional plot.
Inputs:
- chr: [string] chromosome of desired plotting interval
- start: [int] start position of desired plotting interval
- end: [int] end position of desired plotting interval
- variant_data: [data frame] variant-level summary statistics organized with one row per variant, including columns as follows.
- variant_chr_column: [string] column name within variant_data of chromosome values for each variant
- variant_pos_column: [string] column name within variant_data of position values for each variant
- variant_y_column: [string] column name within variant_data of p-values values for each variant
See vignettes/