The qpgraph
Description
The "qpGraph"
class is the class to store and manipulate
qorder (partial) correlation graphs, or qpgraphs for short. See
Castelo and Roverato (2006, 2009) for a mathematical and statistical
definition of a qpgraph.
In earlier versions 1.x of the qpgraph
package there
was a function called qpGraph()
to obtain a qpgraph from a
matrix of nonrejection rates. This function, as it was written,
has been deprecated and replaced by this class and corresponding
constructor methods of the same name. The main difference with respect
to earlier 1.x versions is that the argument threshold
is now
called epsilon
, the argument return.type
has been
removed and the current version returns an object of this class
qpGraph
described in this manual page.
Arguments
epsilon 
maximum cutoff value met by the edges present in the qpgraph. 
topPairs 
number of edges from the top of the ranking, defined by the
nonrejection rates in 
pairup.i 
subset of vertices to pair up with subset 
pairup.j 
subset of vertices to pair up with subset 
q 
qorder employed to derive the input matrix of nonrejection rates

n 
when the input matrix of nonrejection rates 
Objects from the Class
Objects can be created by calls of the form qpGraph(nrrMatrix, ...)
corresponding to constructor methods that take as input a matrix of
nonrejection rates, calculated with qpNrr
.
Slots
p
:number of vertices, in onetoone correspondence with random variables.
q
:order of the qpgraph, always smaller than
p2
.n
:when the qpgraph has been estimated from data, this is the number of observations in the data set, which must be larger than
q+2
.epsilon
:maximum cutoff value of the nonrejection rate met by the edges that are present in the qpgraph.
g
:undirected graph structure of the qpgraph stored as a
graphBAMclass
object.
Methods
qpGraph(nrrMatrix, ...)
Constructor method where
nrrMatrix
is a matrix of nonrejection rates.show(object)
Method to display some bits of information about the qpgraph stored in the input argument
object
.
Author(s)
R. Castelo
References
Castelo, R. and Roverato, A. A robust procedure for Gaussian graphical model search from microarray data with p larger than n, J. Mach. Learn. Res., 7:26212650, 2006.
Castelo, R. and Roverato, A. Reverse engineering molecular regulatory networks from microarray data with qpgraphs. J. Comp. Biol., 16(2):213227, 2009.
See Also
qpNrr