networkPlot: Gene set network plot

In piano: Platform for integrative analysis of omics data

Description

Draws a network with gene sets as nodes and the thickness of the edges correlating to the number of shared genes. The gene set significance is visualized as color intensities. Gives an overview of the influence of overlap on significant gene sets.

Usage

networkPlot(gsaRes, class, direction, adjusted = FALSE,
  significance = 0.001, geneSets = NULL, overlap = 1, lay = 1,
  label = "names", cexLabel = 0.9, ncharLabel = 25, cexLegend = 1,
  nodeSize = c(10, 40), edgeWidth = c(1, 15), edgeColor = NULL,
  scoreColors = NULL, main)

Arguments

gsaRes	an object of class GSAres, as returned from runGSA() or an object returned from runGSAhyper().
class	a character string determining the p-values of which directionality class that should be used as significance information for the plot. Can be one of "distinct", "mixed", "non". Has to be "non" if the result from runGSAhyper() is used.
direction	a character string giving the direction of regulation, can be either "up", "down" or "both" (for class="distinct" only).
adjusted	a logical, if adjusted p-values should be used, or not. Note that if runGSA was run with the argument adjMethod="none", the adjusted p-values will be equal to the original p-values.
significance	the significance cut-off that determines which gene sets are included in the plot. Defaults to 0.001.
geneSets	a character vector of gene set names, to be included in the plot. Defaults to NULL, but if given, the argument significance will not be used.
overlap	a positive numerical. Determines the smallest number of sharing genes between two gene-sets that is needed in order to draw a line/edge between the gene-sets. Defaults to 1.
lay	a numerical between 1-5, or a layout function (see layout in the igraph package). 1-5 sets the layout to one of the five default layout for the network plot.
label	a character string, either "names" ,"numbers", "numbersAndSizes" or "namesAndSizes", determining the labels used for the nodes. The names are the gene set names, numbers is an arbritary numbered list of the gene sets used in the plot connected to the named list returned by the funtion. Sizes are the gene set sizes, e.g. the number of genes.
cexLabel	the text size of the node labels.
ncharLabel	the number of characters to include in the node labels.
cexLegend	the text size of the legend.
nodeSize	a numerical vector of length 2 giving the maximum and minimum node sizes. The node size represents the size of the gene set, and all values will be scaled to the given interval.
edgeWidth	a numerical vector of length 2 giving the maximum and minimum edge widths. The edge width represents the number of shared genes between two gene sets, and all values will be scaled to the given interval.
edgeColor	a character vector giving the colors to use for increasing edge width. Can also be set to a single color. Defaults to a gray-scale.
scoreColors	a character vector giving the colors from which the gradient used for node coloring will be created. In the case of class="distinct" and direction="both" the first half of the vector will be used for the up-regulated gene sets and the second part will be used for the down-regulated gene sets.
main	an optional character vector setting the title of the plot.

Details

In the case of class="distinct" and direction="both", the distinct directional p-values (pDistinctDirUp and pDistinctDirDn, see runGSA) will be used in combination. Using the geneSets and lay arguments, multiple comparative plots (i.e. with the same layout) can be drawn, based for instance on the output gene set list from other network plots with different directionality classes.

Value

Returns a list with two components: geneSets containing the names and numbers of the gene sets in the plot, and layout, containing the saved layout of the plot, which can be passed back to the lay argument in order to draw a subsequent plot with the same layout.

Author(s)

Leif Varemo [e-mail] and Intawat Nookaew [e-mail]

See Also

piano, runGSA, GSAheatmap, networkPlot2, exploreGSAres, layout

Examples

   # Load example input data to GSA:
   data("gsa_input")
   
   # Load gene set collection:
   gsc <- loadGSC(gsa_input$gsc)
      
   # Run gene set analysis:
   gsares <- runGSA(geneLevelStats=gsa_input$pvals , directions=gsa_input$directions, 
                    gsc=gsc, nPerm=500)
      
   # Network plot:
   networkPlot(gsares,class="non",significance=0.01)
   
   # Use circular layout and save the layout:
   nw <- networkPlot(gsares,class="non",significance=0.01,lay=5)
   
   # Use the saved layout to overlay the distinct-directional p-values for easy comparison.
   # Note that the gene sets are now not selected based on a significance cutoff, but from a list:
   networkPlot(gsares,class="distinct",direction="both",lay=nw$layout,geneSets=nw$geneSets)

piano documentation built on Nov. 8, 2020, 6:27 p.m.