| Title: | Biclustering Analysis and Results Exploration |
|---|---|
| Description: | Biclustering Analysis and Results Exploration. |
| Authors: | Pierre Gestraud |
| Maintainer: | Pierre Gestraud <[email protected]> |
| License: | GPL-2 |
| Version: | 1.65.0 |
| Built: | 2024-11-29 04:09:36 UTC |
| Source: | https://github.com/bioc/BicARE |
Biclustering Analysis and Results Exploration
| Package: | BicARE |
| Version: | 0.1.0 |
| Date: | 2008-06-05 |
| Depends: | R (>= 1.8.0), Biobase, multtest, GSEABase |
| License: | GPL |
| biocViews: | Microarray, Transcription, Statistics, Clustering |
| URL: | http://bioinfo.curie.fr |
| Packaged: | Tue Aug 19 14:21:38 2008; pgestraud |
| Built: | R 2.8.0; i686-pc-linux-gnu; 2008-08-19 14:39:57; unix |
Index:
FLOC Performs the FLOC algorithm
bicluster Extract a bicluster
makeReport Export the results as html files
residue Residue of a matrix
sample.bicData Example data set for BicARE
sample.biclustering Example biclustering object
testAnnot Find samples annotations over-represented in
biclusters
testSet Find gene sets that are enriched in a bicluster
Further information is available in the following vignettes:
BicARE |
BicARE (source, pdf) |
Pierre Gestraud
Maintainer: Pierre Gestraud , <[email protected]>
Extract a bicluster from an object of class biclustering
bicluster(biclustering, k, graph=TRUE)bicluster(biclustering, k, graph=TRUE)
biclustering |
an object of class "biclustering" created by
function |
k |
the number of the bicluster considered in the "biclustering" object |
graph |
boolean, indicating whether the graph should be plotted or not |
Returns the bicluster as a matrix with the genes on rows and the samples on columns. Result matrix is of class "bicluster". The "graph" option allows to plot the expression profiles of the genes across the conditions in the bicluster.
Pierre Gestraud
### extract the first bicluster data(sample.biclustering) sample.biclustering bic <- bicluster(sample.biclustering, 1, graph=TRUE) plot(bic)### extract the first bicluster data(sample.biclustering) sample.biclustering bic <- bicluster(sample.biclustering, 1, graph=TRUE) plot(bic)
Find a given number of biclusters using the a modified version of the FLOC algorithm.
FLOC(Data, k = 20, pGene = 0.5, pSample=pGene, r = NULL, N = 8, M = 6, t = 500, blocGene = NULL, blocSample = NULL)FLOC(Data, k = 20, pGene = 0.5, pSample=pGene, r = NULL, N = 8, M = 6, t = 500, blocGene = NULL, blocSample = NULL)
Data |
an
|
k |
the number of biclusters searched |
pGene |
genes initial probability of membership to the biclusters |
pSample |
samples initial probability of membership to the biclusters |
r |
the residue threshold |
N |
minimal number of gene per bicluster |
M |
minimal number of conditions per bicluster |
t |
number of iterations |
blocGene |
a matrix indicating the directed initialisation for the genes (see details) |
blocSample |
a matrix indicating the directed initialisation for the conditions (see details) |
This biclustering algorithm is based on the FLOC algorithm (FLexible
Overlapped biClustering) defined by Yang et al. (see references). It
can discover a set of k, possibly overlapping, biclusters. If
r is set to
NULL, the residue threshold used in the analysis is the residue of
Data divided by 10.
blocGene and blocSample are matrix of 0 and 1 with the rows
representing the features (gene or samples) and the columns the
biclusters. A 1 on line i and column j indicates that the feature i
(gene or sample) will be include in the bicluster j during the
initialisation step and will not be removed from it during the analysis. If the number of columns in these matrices is different from the number of bicluster searched, k is set to the maximal value
of these two.
See bicluster to extract a bicluster from the biclustering result.
Returns an object of class 'biclustering', a list containing at least :
Call |
the matched call. |
ExpressionSet |
the data used |
param |
a data.frame with the algorithm parameters |
bicRow |
a matrix of boolean indicating the belonging of the genes to the biclusters |
bicCol |
the same as for bicRow but for the conditions |
mat.resvol.bic |
a matrix describing the biclusters |
Pierre Gestraud ([email protected])
J. Yang, H. Wang, W. Wang, and P.S. Yu. An improved biclustering method for analyzing gene expression. International Journal on Artificial Intelligence Tools, 14(5):771-789, 2005
data(sample.bicData) ## subset of sample.ExpressionSet from Biobase residue(sample.bicData) ## 0.3401921 resBic <- FLOC(sample.bicData, k=10, pGene=0.5,r=0.05,N=8,M=10,t=500) resBic ## initialising samples of 2 biclusters iniSample <- matrix(0, ncol=2, nrow=26) ## first bicluster initialised around Female cases iniSample[pData(sample.bicData)$sex=="Female",1] <- 1 ## second bicluster initialised around control cases iniSample[pData(sample.bicData)$type=="Control",2] <- 1 resBic <- FLOC(sample.bicData, k=10, pGene=0.5, r=0.05, N=8, M=10, t=500, blocSample=iniSample) resBicdata(sample.bicData) ## subset of sample.ExpressionSet from Biobase residue(sample.bicData) ## 0.3401921 resBic <- FLOC(sample.bicData, k=10, pGene=0.5,r=0.05,N=8,M=10,t=500) resBic ## initialising samples of 2 biclusters iniSample <- matrix(0, ncol=2, nrow=26) ## first bicluster initialised around Female cases iniSample[pData(sample.bicData)$sex=="Female",1] <- 1 ## second bicluster initialised around control cases iniSample[pData(sample.bicData)$type=="Control",2] <- 1 resBic <- FLOC(sample.bicData, k=10, pGene=0.5, r=0.05, N=8, M=10, t=500, blocSample=iniSample) resBic
Creates a directory with html files containing the biclustering results.
makeReport(dirPath, dirName, resBic, browse=TRUE)makeReport(dirPath, dirName, resBic, browse=TRUE)
dirPath |
path to the directory |
dirName |
the name of the directory where the report will be created |
resBic |
a biclustering result |
browse |
logical. If TRUE the web browser will be opened |
makeReport produces a html report of biclustering results in a new directory named dirName.
If the browse argument is set to TRUE the web browser will be opened on the
"home.html" file.
Make sure to have rights to create the result directory.
Pierre Gestraud [email protected]
data(sample.biclustering) dirPath <- getwd() ## report created in the current working directory dirName <- "test" makeReport(dirPath, dirName, sample.biclustering, browse=FALSE)data(sample.biclustering) dirPath <- getwd() ## report created in the current working directory dirName <- "test" makeReport(dirPath, dirName, sample.biclustering, browse=FALSE)
Returns the residue of a matrix.
residue(Data)residue(Data)
Data |
an
|
This function computes the residue of a matrix as defined by Yang et al (see references).
Pierre Gestraud
J. Yang, H. Wang, W. Wang, and P.S. Yu. An improved biclustering method for analyzing gene expression. International Journal on Artificial Intelligence Tools, 14(5):771-789, 2005
data(sample.bicData) residue(sample.bicData)data(sample.bicData) residue(sample.bicData)
A subset of sample.ExpressionSet from package Biobase. The data for 26 cases, labeled A to Z and 350 genes. Each case has three covariates: sex (male/female), type (case/control) and score (testing score).
sample.bicDatasample.bicData
An ExpressionSet
A biclustering object created by the FLOC
function on the sample.bicData with the following options
: k=10, pGene = 0.3, pSample = 0.5,
r = 0.025, N = 8, M = 8, t = 1000.
sample.biclusteringsample.biclustering
a biclustering object
Characterisation of the biclusters in term of over-representation of sample covariates.
testAnnot(resBic, annot=NULL, covariates="all")testAnnot(resBic, annot=NULL, covariates="all")
resBic |
a biclustering result from FLOC |
annot |
annotation matrix, default value is set to NULL, then phenoData of the ExpressionSet is used |
covariates |
the names of the covariates that should be tested, default value is set to "all" |
For each bicluster and each covariate a chi-squarred test is performed to test the adequation between the distribution of the levels of the covariates in the bicluster and in the original dataset.
Multiple testing correction is performed by the Benjamini-Yekutieli procedure. The residuals of the tests indicate if the level is over or down represented in the bicluster.
Due to the amount of results it is advised to use the makeReport function to get a html report.
A biclustering object containing resBic and updated with
the results of the tests in resBic$covar.
The results are presented as a list with :
covar |
the samples covariates tested |
pvalues |
a matrix with the p-values of the tests |
adjpvalues |
a matrix with the p-values adjusted by the Benjamini Yekutieli procedure |
index |
a list of matrices with the numbers of each level in each bicluster |
residuals |
a list of matrices with the residuals of the tests for each modality in each bicluster |
Pierre Gestraud
data(sample.biclustering) resBic <- testAnnot(sample.biclustering, annot=NULL, covariates=c("sex", "type"))data(sample.biclustering) resBic <- testAnnot(sample.biclustering, annot=NULL, covariates=c("sex", "type"))
Test of the over-representation of gene sets in the biclusters
testSet(resBic, geneSetCol)testSet(resBic, geneSetCol)
resBic |
a |
geneSetCol |
The over-representation of a gene set in a bicluster is evaluated by an hypergeometric test.
The genes identifiers of the gene sets will automatically be mapped to the same as those used in the data.
Due to the amount of results it is advised to use the makeReport function to get a html report.
A biclustering object containing resBic and updated with
the results of the tests in resBic$geneSet.
The results are presented as a list with :
GeneSetCollection |
the GeneSetCollection used |
pvalues |
a matrix containing the pvalues of the tests for each geneSet and each bicluster |
adjpvalue |
a matrix containing the p-values adjusted by the Benjamini Yekutieli procedure |
Pierre Gestraud [email protected]
data(sample.biclustering) gss <- GeneSetCollection(sample.biclustering$ExpressionSet[1:50,], setType=GOCollection()) resBic <- testSet(sample.biclustering, gss)data(sample.biclustering) gss <- GeneSetCollection(sample.biclustering$ExpressionSet[1:50,], setType=GOCollection()) resBic <- testSet(sample.biclustering, gss)