Type: Package
Title: Discovery of Candidate Disease Genes Using Trans-Regulatory Effects
Version: 0.1.0
Description: Implements the DANDELION method for prioritizing candidate disease-related proximal genes by integrating trans-regulatory association p-values and gene-level trait association p-values. The statistical testing framework builds on the divide-aggregate composite-null test described by Liu et al. (2022) <doi:10.1080/01621459.2021.1914634>. The biological application and SNP/gene-based prioritization workflow are motivated by Salamone et al. (under review), "Leveraging trans-gene regulation prioritizes central genes and pathways in asthma". The package provides functions for identifying distal-proximal gene pairs, organizing significant pairs into genomic loci, and visualizing resulting gene networks.
License: GPL-3
Encoding: UTF-8
RoxygenNote: 7.3.3
Imports: igraph
Suggests: qvalue
Additional_repositories: https://bioconductor.org/packages/release/bioc
NeedsCompilation: no
Packaged: 2026-04-30 07:05:09 UTC; px
Author: Peixin Tian [aut, cre]
Maintainer: Peixin Tian <pxtian@connect.hku.hk>
Repository: CRAN
Date/Publication: 2026-05-04 19:20:09 UTC

Organize DANDELION Gene-Gene Results

Description

Cleans significant DANDELION pairs when the exposure side contains genes and merges nearby exposure genes into genomic loci.

Usage

calc_pair.gene(
  mat.sig,
  mat.p,
  p.wes,
  gene1,
  ref.table.keep,
  eta.wgs = 1e-05,
  verbose = FALSE
)

Arguments

mat.sig

Matrix encoding significant pairs, returned by med_gene().

mat.p

Matrix of DANDELION p-values, returned by med_gene().

p.wes

Named numeric vector of gene-level trait association p-values.

gene1

Candidate exposure genes returned by med_gene().

ref.table.keep

Gene annotation data frame after filtering. Required columns are gene_name, Chromosome, start, and end.

eta.wgs

Significance threshold for WES genes. Default is 1e-5.

verbose

Logical. If TRUE, progress messages are printed. Default is FALSE.

Value

A list containing pairs_dact, gene.pair, sig_gene2, and non_sig.gene2.

Organize DANDELION SNP-Gene Results

Description

Cleans significant DANDELION pairs when the exposure side contains SNPs. SNPs are mapped to cis genes using uniq_snp first. If a SNP is not present in uniq_snp, the function optionally maps it to the nearest or overlapping gene using SNP.ref and ref.table.keep.

Usage

calc_pair.snp(
  mat.sig,
  mat.p,
  p.wes,
  gene1,
  uniq_snp,
  ref.table.keep,
  eta.wgs = 1e-05,
  SNP.ref = NULL,
  verbose = FALSE
)

Arguments

mat.sig

Matrix encoding significant pairs, returned by med_gene().

mat.p

Matrix of DANDELION p-values, returned by med_gene().

p.wes

Named numeric vector of gene-level trait association p-values.

gene1

Candidate SNPs returned by med_gene().

uniq_snp

Data frame containing known SNP-to-cis-gene mapping. Required columns are SNP and GeneSymbol.

ref.table.keep

Gene annotation data frame after filtering. Required columns are gene_name, Chromosome, start, and end.

eta.wgs

Significance threshold for WES genes. Default is 1e-5.

SNP.ref

Optional SNP annotation data frame with columns SNP, SNPPos, and SNPChr. Used only for SNPs not mapped by uniq_snp.

verbose

Logical. If TRUE, progress messages are printed. Default is FALSE.

Value

A list containing pairs_dact, gene.pair, sig_gene2, and non_sig.gene2.

Generate DANDELION Network Figure

Description

Generates a network plot from DANDELION gene-pair results.

Usage

gen_fig(gene.pair, p.wes, eta.wgs = 1e-05, pic_dir)

Arguments

gene.pair

Data frame of identified gene pairs, returned by calc_pair.gene() or calc_pair.snp(). Must contain columns region and gene2.

p.wes

Named numeric vector of gene-level trait association p-values.

eta.wgs

Significance threshold for WES genes. Default is 1e-5.

pic_dir

Directory to save the figure.

Value

Invisibly returns the path to the saved PDF file.

Apply DANDELION to Identify Candidate Disease Proximal Genes

Description

Applies the DANDELION procedure to integrate trans-regulatory association p-values and gene-level trait association p-values. The exposure side can be either distal genes or SNPs. For clarity, exposures are referred to as gene1, and candidate disease proximal genes are referred to as gene2.

Usage

med_gene(
  p.trans,
  p.wes,
  ref.table,
  gene1.list,
  target.fdr = 0.1,
  dist = 5e+06,
  gene1.type = c("SNP", "Gene"),
  SNP.ref = NULL,
  n.cores = 1,
  verbose = FALSE
)

Arguments

p.trans

A numeric matrix of trans-regulatory association p-values. Rows are candidate disease proximal genes (gene2), and columns are exposures (gene1), either distal genes or SNPs. Row names and column names must be provided. (e.g., from trans-eQTL studies like eQTLGen)

p.wes

A named numeric vector of gene-level trait association p-values (e.g., from WES burden tests or GWAS-based gene-level tests). Note: Ensure that names (Gene Symbols) match the row names of p.trans

ref.table

A data frame containing gene annotation and genomic positions. Required columns are gene_name, type, Chromosome, start, and end. Chromosome should use the format chr1, chr2, etc.

gene1.list

A character vector of candidate exposures to analyze. These values must be a subset of colnames(p.trans).

target.fdr

False Discovery Rate threshold. DANDELION uses this to decide which gene pairs are statistically significant. Default is 0.1.

dist

The cis-window size (in base pairs). Genes within this distance from the exposure will be excluded to focus on true distal (trans) effects. Default is 5e6.

gene1.type

Exposure type. Must be either "SNP" or "Gene".

SNP.ref

Optional SNP annotation data frame used when gene1.type = "SNP". Required columns are SNP, SNPPos, and SNPChr. SNPChr can be 1 or chr1; both formats are accepted.

n.cores

Number of cores for parallel execution. On non-Windows systems, values larger than 1 use parallel::mclapply(). On Windows, the function falls back to single-core execution.

verbose

Logical. If TRUE, progress messages are printed. Default is FALSE.

Value

A list with three elements: gene1, the analyzed exposures with at least one valid DANDELION result; mat.sig, a matrix encoding significant gene1-gene2 pairs; and mat.p, a matrix of DANDELION p-values.

Examples

set.seed(1)
p.trans <- matrix(runif(60, 0.001, 0.9), nrow = 12, ncol = 5)
rownames(p.trans) <- paste0("G", 1:12)
colnames(p.trans) <- paste0("E", 1:5)
p.wes <- runif(12, 0.001, 0.9)
names(p.wes) <- paste0("G", 1:12)
ref.table <- data.frame(
  gene_name = c(paste0("G", 1:12), paste0("E", 1:5)),
  type = "protein_coding",
  Chromosome = "chr1",
  start = seq_len(17) * 1e7,
  end = seq_len(17) * 1e7 + 1000
)
res <- med_gene(
  p.trans = p.trans,
  p.wes = p.wes,
  ref.table = ref.table,
  gene1.list = colnames(p.trans),
  gene1.type = "Gene"
)

Estimate the Non-null Proportion

Description

Internal helper used by DANDELION.

Usage

nonnullPropEst(x, u, sigma)

Arguments

x

Numeric vector.

u

Null mean.

sigma

Null standard deviation.

Value

Estimated non-null proportion.

Print DANDELION pair results

Description

Print DANDELION pair results

Usage

## S3 method for class 'dandelion_pairs'
print(x, ...)

Arguments

x

A dandelion_pairs object returned by calc_pair.gene() or calc_pair.snp().

...

Additional arguments, currently unused.

Value

Invisibly returns x.

Print a DANDELION result object

Description

Print a DANDELION result object

Usage

## S3 method for class 'dandelion_result'
print(x, ...)

Arguments

x

A dandelion_result object returned by med_gene().

...

Additional arguments, currently unused.

Value

Invisibly returns x.