get_GI_network: Get gene-interaction network.

In hclimente/martini: GWAS Incorporating Networks

Get gene-interaction network.

Description

Creates a network of SNPs where each SNP is connected as in the GM network and, in addition, to all the other SNPs pertaining to any interactor of the gene it is mapped to. Corresponds to the gene-interaction (GI) network described by Azencott et al.

Usage

get_GI_network(
  gwas,
  organism = 9606,
  snpMapping = snp2ensembl(gwas, organism),
  ppi = get_gxg("biogrid", organism, flush),
  col_ppi = c("gene1", "gene2"),
  col_genes = c("snp", "gene"),
  flush = FALSE
)

Arguments

gwas	A SnpMatrix object with the GWAS information.
organism	Tax ID of the studied organism. The default is 9606 (human).
snpMapping	A data.frame informing how SNPs map to genes. It contains minimum two columns: SNP id and a gene it maps to. Each row corresponds to one gene-SNP mapping. Unless column names are specified using col_genes, involved columns must be named 'snp' and 'gene'.
ppi	A data.frame describing protein-protein interactions with at least two colums. Gene ids must be the contained in snpMapping. Unless column names are specified using col_ppi, involved columns must be named gene1 and gene2.
col_ppi	Optional, length-2 character vector with the names of the two columns involving the protein-protein interactions.
col_genes	Optional, length-2 character vector with the names of the two columns involving the SNP-gene mapping. The first element is the column of the SNP, and the second is the column of the gene.
flush	Remove cached results? Boolean value.

Value

An igraph network of the GI network of the SNPs.

References

Azencott, C. A., Grimm, D., Sugiyama, M., Kawahara, Y., & Borgwardt, K. M. (2013). Efficient network-guided multi-locus association mapping with graph cuts. Bioinformatics, 29(13), 171-179. https://doi.org/10.1093/bioinformatics/btt238

Examples

get_GI_network(minigwas, snpMapping = minisnpMapping, ppi = minippi)

hclimente/martini documentation built on Feb. 26, 2024, 6:23 p.m.