Retrieve pathway topology as weighted adjacency matrix

Source: R/pathwayTopology.R

Retrieve pathway topology matrices and convert them to normalized weighted directed adjacency matrices describing gene signaling networks.

retrieve_topology(
  database = c("kegg", "pathbank", "wikipathways", "reactome", "panther", "pharmgkb",
    "smpdb"),
  species = c("hsapiens", "athaliana", "btaurus", "celegans", "cfamiliaris",
    "dmelanogaster", "drerio", "ecoli", "ggallus", "mmusculus", "rnorvegicus",
    "scerevisiae", "sscrofa", "xlaevis"),
  keyword = NULL,
  beta = NULL
)

Arguments

database

A character vector of supported databases.

species

One of the supported species,

keyword

Optional. Areas of interests as a character vector.

beta

Optional. A named numeric vector of weights to be assigned to each type of gene/protein relation type.See details for more information.

Value

A list where each element is a matrix corresponding to a pathway

Details

This function takes pathway topology information retrieved using graphite and converts these to normalized weighted directed adjacency matrices describing the gene signaling network, which can be used to compute gene-wise and pathway-level perturbation score through the scoring algorithm derived from the SPIA algorithm. See cited document for more details.

The database parameter may specify multiple databases but only one species can provided to the species parameter.

Users can provide areas of interests as keywords, which will be matched to pathway names from chosen databases to subset pathways. Cases will be ignored in string matchings.

The beta parameter specifies weights to be assigned to each type of gene-gene interaction. It should be a named numeric vector of length 25, whose names must be: c("activation","compound","binding/association", "expression","inhibition","activation_phosphorylation","phosphorylation", "indirect","inhibition_phosphorylation","dephosphorylation_inhibition", "dissociation","dephosphorylation","activation_dephosphorylation", "state","activation_indirect","inhibition_ubiquination","ubiquination", "expression_indirect","indirect_inhibition","repression", "binding/association_phosphorylation","dissociation_phosphorylation", "indirect_phosphorylation").

If unspecified, beta will be set as an integer vector with: a) values of 1 for interactions which match 'expression' or 'activation'; b) values of -1 for interactions which match 'repression' or 'inhibition'; and c) 0 elsewhere. The converted weighted adjacent matrices will be stored in a list. We recommend users to store the returned list as a file so this step only needs to be performed once for each database.

References

Tarca AL, Draghici S, Khatri P, Hassan SS, Mittal P, Kim JS, Kim CJ, Kusanovic JP, Romero R. A novel signaling pathway impact analysis. Bioinformatics. 2009 Jan 1;25(1):75-82. Sales, G., Calura, E., Cavalieri, D. et al. graphite - a Bioconductor package to convert pathway topology to gene network. BMC Bioinformatics 13, 20 (2012).

Examples

# \donttest{
# retrieve pathway topology matrices of all KEGG pathway
gsTopology <- retrieve_topology(database = "kegg", species = "hsapiens")

# If only interested in selected pathways, specify the areas of interest as
# keywords
gsTopology <- retrieve_topology(database = "kegg",
keyword = c("metabolism", "signaling"), species = "hsapiens")
# }