analyzePairs: Finds all mutually exclusive pairs in a dataset

In cbg-ethz/TiMEx: Finding Mutually Exclusive Groups of Alterations in Cancer Datasets

Description

analyzePairs performs step 1 of the TiMEx procedure: tests all gene pairs for mutual exclusivity. It returns a complex list, including parameter estimates, pairwise p-values and intensities of mutual exclusivity, log likelihoods, and others.

Usage

analyzePairs(mat)

Arguments

mat	binary alteration matrix, with rows representing patients and columns representing genes

Details

In the first step, the TiMEx procedure for identifying mutually exclusive groups of alterations tests all pairs for mutual exclusivity. The data corresponding to each pair of genes is therefore fitted to both the Null (Conditional Independence) and the Mutual Exclusivity models. Parameters under the two models are estimated, and, since they are nested, a likelihood ratio test is performed between the corresponding log likelihoods, in order to test whether mu (the intensity of mutual exclusivity) is different from 0. For more details on the TiMEx procedure, as well as on the underlying mathematical model, see Constantinescu et al.: TiMEx: A Waiting Time Model for Mutually Exclusive Cancer Alterations. Bioinformatics (2015).

The output list contains exhaustive information on the pairwise testing of genes, including parameter estimates, pairwise p-values and intensities of mutual exclusivity, log likelihoods, and others.

This function displays progress messages indicating the gene that is currently being tested against the remaining genes.

Value

List consisting of a set of matrices, all of dimension n x n (n being the number of genes). Element (i,j) of each matrix corresponds to the pairwise test between genes i and j:

• lamiEstNull matrix with rate estimates (lambda) of the waiting time of one gene (gene i) under the Null model.

• lamjEstNull matrix with rate estimates (lambda) of the waiting time of the other gene (gene j) under the Null model.

• lamiEstME matrix with rate estimates (lambda) of the waiting time of one gene (gene i) under the Mutual Exclusivity model.

• lamjEstME matrix with rate estimates (lambda) of the waiting time of the other gene (gene j) under the Mutual Exclusivity model.

• likeNull matrix with log likelihoods under the Null model.

• likeAlt matrix with log likelihoods under the Mutual Exclusivity model.

• LRT matrix with log likelihood ratio test (LRT) statistics.

• pvalueLRTCorrectSym list of the following symmetric matrices:

  • bonferroni bonferroni corrected p-values corresponding to the LRT statistic.

  • bonferroni fdr corrected p-values corresponding to the LRT statistic.

  • uncorrected uncorrected p-values corresponding to the LRT statistic.

• muEstSym symmetric matrix with estimated pairwise mu (intensity of mutual exclusivity)

Author(s)

Simona Cristea, [e-mail]

References

Constantinescu et al.: TiMEx: A Waiting Time Model for Mutually Exclusive Cancer Alterations. Bioinformatics (2015).

See Also

doMaxCliques for step 2 of the TiMEx procedure; findSignifCliques for step 3 of the TiMEx procedure; the wrapper function TiMEx for combining these three steps, and identifying mutually exclusive groups in a binary dataset with the TiMEx model.

Examples

# Test all pairs from the ovarian dataset for mutual exclusivity (takes 
# approximately 5 minutes)
data(ovarian)
ovarianPairs<-analyzePairs(ovarian)

cbg-ethz/TiMEx documentation built on May 13, 2019, 1:50 p.m.