Introduction

pleior is an R package for analyzing pleiotropy in genome-wide association studies (GWAS).

Here, we provide a basic workflow for using pleior and how to use its core functions to load, preprocess, and analyze GWAS data for pleiotropic effects.

Setup

First, load the pleior package and the example dataset.

library(pleior)
data(gwas_subset)
head(gwas_subset)
#>         SNPS          MAPPED_TRAIT PVALUE_MLOG CHR_ID   CHR_POS
#>       <char>                <char>       <num> <char>    <char>
#> 1:  rs814573     Alzheimer disease   672.69900     19  44908684
#> 2:  rs814573 myocardial infarction    15.00000     19  44908684
#> 3:    rs7412     Alzheimer disease   122.39790     19  44919689
#> 4:    rs7412       LDL cholesterol  9629.00000     19  44919689
#> 5: rs2817462    memory performance     6.30103      6 156588550
#> 6: rs2817462  tyrosine measurement     9.69897      6 156588550

Workflow

Step 1: Load GWAS data

Although we use the example dataset here, you can load your own GWAS summary statistics.

gwas_data <- gwas_subset

Step 2: Preprocess GWAS data

Filter the data to retain significant associations.

gwas_clean <- preprocess_gwas(gwas_data, pvalue_threshold = 5e-8)
head(gwas_clean)
#>          SNPS          MAPPED_TRAIT PVALUE_MLOG CHR_ID   CHR_POS
#>        <char>                <char>       <num> <char>    <char>
#> 1:   rs814573     Alzheimer disease   672.69900     19  44908684
#> 2:   rs814573 myocardial infarction    15.00000     19  44908684
#> 3:     rs7412     Alzheimer disease   122.39790     19  44919689
#> 4:     rs7412       LDL cholesterol  9629.00000     19  44919689
#> 5:  rs2817462  tyrosine measurement     9.69897      6 156588550
#> 6: rs10182181       body mass index    29.69897      2  24927427

Step 3: Detect pleiotropic SNPs

Identify SNPs associated with multiple traits.

pleio_results <- detect_pleiotropy(gwas_clean, traits = c("Alzheimer disease", "myocardial infarction", "LDL cholesterol"))
head(pleio_results)
#>        SNPS N_TRAITS                                  TRAITS
#>      <char>    <int>                                  <char>
#> 1:   rs7412        2       Alzheimer disease;LDL cholesterol
#> 2:   rs7412        2       Alzheimer disease;LDL cholesterol
#> 3: rs814573        2 Alzheimer disease;myocardial infarction
#> 4: rs814573        2 Alzheimer disease;myocardial infarction
#>             MAPPED_TRAIT PVALUE_MLOG CHR_ID  CHR_POS
#>                   <char>       <num> <char>   <char>
#> 1:     Alzheimer disease    122.3979     19 44919689
#> 2:       LDL cholesterol   9629.0000     19 44919689
#> 3:     Alzheimer disease    672.6990     19 44908684
#> 4: myocardial infarction     15.0000     19 44908684

Step 4: Visualize results

Create a Manhattan plot to visualize pleiotropic SNPs.

plot_pleiotropy_manhattan(pleio_results, highlight_snp = "rs814573")

For more advanced features, such as functional annotation or interactive visualization, refer to the package documentation.