gMCP (version 0.8-17)

graphGUI: Graphical User Interface for graphical described multiple comparison procedures

Description

Starts a graphical user interface for the creation/modification of directed weighted graphs and applying graphical described multiple comparison procedures.

Usage

graphGUI(
  graph = "createdGraph",
  pvalues = numeric(0),
  grid = 0,
  debug = FALSE,
  experimentalFeatures = FALSE,
  envir = globalenv()
)

Value

The function itself returns NULL. But with the GUI a graph can be created or edited that will be available in R under the specified variable name after saving in the specified environment.

Arguments

graph

Either a variable name for the graph, given as a character string. (If it is not a syntactically valid name, make.names is called to change it to a valid one.) Or an object of class graphMCP. If the object is modified (even just by updating the class definition or arranging the nodes) it will be saved in the specified environment (default is the global environment).

pvalues

Numeric value that optionally specifies the p-values.

grid

Positive integer that sets the grid size for easier placement of nodes. (Therefore grid size 1 allows unrestricted placement and disables the grid.) The default grid=0 uses the last used grid value or if the GUI is started the first time a value of 50.

debug

Logical. If TRUE debug output is printed to the R console.

experimentalFeatures

Logical. If TRUE some unfinished / insufficiently tested experimental features are available in the GUI.

envir

Environment where the object graph is located and/or it should be saved (default is the global environment).

Author

Kornelius Rohmeyer

Details

See the vignette of this package for further details, since describing a GUI interface is better done with a lot of nice pictures.

The GUI can save result files if asked to, can look for a new version on CRAN (if this behaviour has been approved by the user), will change the random seed in the R session if this is specified by the user in the options (default: no) and could send bug reports if an error occurs and the user approves it.

Examples

if (FALSE) {
graphGUI()
pvalues <- c(9.7, 1.5, 0.5, 0.6, 0.4, 0.8, 4)/100 
graphGUI(HommelEtAl2007(), pvalues=pvalues)

x <- new.env()
assign("graph", BonferroniHolm(3), envir=x)
graphGUI("graph", envir=x)}