equivClust: For a given level (2, 3, ...) in a GO ontology (BP, MF or...

In alexsanchezpla/goProfiles: goProfiles: an R package for the statistical analysis of functional profiles

Description

For a given level (2, 3, ...) in a GO ontology (BP, MF or CC), compute the equivalence threshold distance matrix and generate a dendrogram from it.

Usage

equivClust(ontoLevel, onto, geneLists, trace = TRUE, onTheFlyDev = NULL,
  method = "complete", jobName = paste("Equivalence cluster", onto,
  ontoLevel, method, sep = "_"), ylab = "Equivalence threshold distance",
  alpha = 0.05, precis = 0.001, ...)

Arguments

ontoLevel	integer (2, 3, ...) level of a GO ontology where the GO profiles are built
onto	character, GO ontology ("BP", "MF" or "CC") under consideration
geneLists	list of character vectors, each vector stands for the gene names in a given gene set
trace	boolean, the full process must be traced? Defaults to TRUE
onTheFlyDev	character, name of the graphical device where to immediately display the resulting diagram. The appropriate names depend on the operating system. Defaults to NULL and then nothing is displayed
method	character, one of the admissible methods in function hclust. Defaults to "complete"
jobName	character, main plot name, defaults to paste("Equivalence cluster", onto, ontoLevel, method, sep = "_")
ylab	character, label of the vertical axis of the plot, defaults to "Equivalence threshold distance"
alpha	simultaneous nominal significance level for the equivalence tests to be repeteadly performed, defaults to 0.05
precis	numerical precission in the iterative search of the equivalence threshold distances, defaults to 0.001
...	additional arguments to hclust

Details

Do not confuse the threshold distance matrix with the squared distances computed in each equivalence test.

Value

An object of class equivClust, descending from class hclust with some additional attributes:

jobName

The main job name

sub

The graphic subtittle

ylab

The vertical axis label

distMat

The equivalence threshold distance matrix

allComps

A list with some information on all the pairwise equivalence tests: the Euclidean squared distance, its standard error and the corresponding GO profiles

Examples

## Not run: 
data(kidneyGeneLists)
clustMF2 <- equivClust(2, "MF", kidneyGeneLists, orgPackage="org.Hs.eg.db")
plot(clustMF2)
plot(clustMF2, 
     main = "Dendrogram (method = complete)", sub = attr(clustMF2, "sub"),
     ylab = "Equivalence threshold distance")
# With the same data, an UPGMA dendrogram:
equivClust(2, "MF", kidneyGeneLists, method = "average",
orgPackage="org.Hs.eg.db")

## End(Not run)

alexsanchezpla/goProfiles documentation built on May 28, 2019, 4:54 p.m.