Identifies SNPs associated with multiple traits in GWAS data. A SNP is considered pleiotropic if it shows significant associations with two or more distinct traits after applying the significance threshold.
Usage
detect_pleiotropy(
gwas_data,
traits = NULL,
pvalue_threshold = 5e-08,
adjust_method = c("none", "bonferroni", "fdr"),
clump_kb = NULL
)Arguments
- gwas_data
A data.frame containing GWAS summary statistics.
- traits
Character vector. Specific traits to analyze for pleiotropy (default: NULL uses all traits). Matching is exact against individual traits after splitting multi-trait fields.
- pvalue_threshold
Numeric. P-value threshold for significance (default: 5e-8).
- adjust_method
Character. Multiple-testing correction applied to all associations before thresholding: "none" (default), "bonferroni" or "fdr".
- clump_kb
Numeric or NULL. If supplied, co-located SNPs within this distance (kilobases) are clumped to their most significant representative. This is a position-based proxy for LD clumping (default: NULL, no clumping).
Value
A data.frame with pleiotropic SNPs, one row per SNP-trait association, including `N_TRAITS` and `TRAITS` summary columns.
Details
The significance filter is applied *before* traits are counted, so every reported pleiotropic SNP has at least two associations passing the threshold. Multi-trait `MAPPED_TRAIT` fields (comma- or semicolon-separated) are expanded into individual traits.
Examples
data(gwas_subset)
pleio_results <- detect_pleiotropy(gwas_subset)
head(pleio_results)
#> SNPS MAPPED_TRAIT PVALUE_MLOG CHR_ID
#> 1 rs429358 C-reactive protein measurement 1952.69897 19
#> 2 rs429358 high density lipoprotein cholesterol measurement 497.00000 19
#> 3 rs429358 Alzheimer disease 323.00000 19
#> 4 rs429358 total cholesterol measurement 323.00000 19
#> 5 rs429358 low density lipoprotein cholesterol measurement 186.52288 19
#> 6 rs429358 coronary artery disease 42.39794 19
#> CHR_POS N_TRAITS
#> 1 44908684 12
#> 2 44908684 12
#> 3 44908684 12
#> 4 44908684 12
#> 5 44908684 12
#> 6 44908684 12
#> TRAITS
#> 1 Alzheimer disease;C-reactive protein measurement;body height;body mass index;coronary artery disease;high density lipoprotein cholesterol measurement;low density lipoprotein cholesterol measurement;myocardial infarction;systolic blood pressure;total cholesterol measurement;triglyceride measurement;type 2 diabetes mellitus
#> 2 Alzheimer disease;C-reactive protein measurement;body height;body mass index;coronary artery disease;high density lipoprotein cholesterol measurement;low density lipoprotein cholesterol measurement;myocardial infarction;systolic blood pressure;total cholesterol measurement;triglyceride measurement;type 2 diabetes mellitus
#> 3 Alzheimer disease;C-reactive protein measurement;body height;body mass index;coronary artery disease;high density lipoprotein cholesterol measurement;low density lipoprotein cholesterol measurement;myocardial infarction;systolic blood pressure;total cholesterol measurement;triglyceride measurement;type 2 diabetes mellitus
#> 4 Alzheimer disease;C-reactive protein measurement;body height;body mass index;coronary artery disease;high density lipoprotein cholesterol measurement;low density lipoprotein cholesterol measurement;myocardial infarction;systolic blood pressure;total cholesterol measurement;triglyceride measurement;type 2 diabetes mellitus
#> 5 Alzheimer disease;C-reactive protein measurement;body height;body mass index;coronary artery disease;high density lipoprotein cholesterol measurement;low density lipoprotein cholesterol measurement;myocardial infarction;systolic blood pressure;total cholesterol measurement;triglyceride measurement;type 2 diabetes mellitus
#> 6 Alzheimer disease;C-reactive protein measurement;body height;body mass index;coronary artery disease;high density lipoprotein cholesterol measurement;low density lipoprotein cholesterol measurement;myocardial infarction;systolic blood pressure;total cholesterol measurement;triglyceride measurement;type 2 diabetes mellitus
# Analyze specific traits
specific_traits <- c("Alzheimer disease", "myocardial infarction")
pleio_specific <- detect_pleiotropy(gwas_subset, traits = specific_traits)