topologyGSA (version 1.4.5)

clique.mean.test: Comparison of all the cliques of a pathway.

Description

Test the equality of the means in two experimental conditions for all the cliques of a pathway, conditioning on the results of the test on the concentration matrices

Usage

clique.mean.test(y1,y2,dag,alpha,perm.num=1000,paired=FALSE)

Arguments

y1

a matrix with n1 individuals (rows) in the first experimental condition and p genes (columns).

y2

a matrix with n2 individuals (rows) in the second experimental condition and p genes (columns). The genes in the two experimental conditions must be the same.

dag

graphNEL object, a directed acyclic graph (DAG) corresponding to the pathway of interest. See package gRbase for more details.

alpha

significance level of the test.

perm.num

number of permutation used by the test.

paired

logical flag. If TRUE statistical tests for paired samples is calculated and the test on the variances is not performed.

Value

p.value

p-value of the test on the mean for each clique.

p.value.var

p-value of the test on the variance for each clique. If paired is TRUE, this is not returned.

cliques

the list of the corresponding cliques.

var.equal

logical, a vector with a TRUE for a significant clique and a FALSE otherwise. If paired is TRUE, this is not returned.

t.value

the observed values of the statistic to test the mean for each clique.

lambda.value

the observed values of the statistic to test the variance for each clique. If paired is TRUE, this is not returned.

graph

the triangularized and moralized graph.

Details

The function tests the equality of the means of each clique of a pathway in two experimental conditions, conditioning on the results of clique.var.test. The graph of a pathway is first converted into a DAG, then the moral graph is obtained and if the latter graph is decomposable then the test is performed on all its cliques. If the moral graph is not decomposable, its triangulated version is obtained and then the test is performed on all its cliques.

The expression data may contain some genes differing from those in the pathway: in such case the function automatically takes the intersection between the two gene sets.

This function requires gRBase and qpgraph packages.

References

Massa, M.S., Chiogna, M., Romualdi, C. (2010). Gene set analysis exploiting the topology of a pathway. BMC Systems Biology, 4:121 http://www.biomedcentral.com/1752-0509/4/121

See Also

pathway.var.test,pathway.mean.test, clique.var.test,

Examples

data(examples)

clique.mean.test(y1, y2, dag_bcell, 0.05)