HTSanalyzeR4cellHTS2: An analysis pipeline for cellHTS2 objects

In HTSanalyzeR: Gene set over-representation, enrichment and network analyses for high-throughput screens

Description

This function writes an html report following a complete analyses of a dataset based on the two classes GSCA (Gene Set Collection Analysis) and NWA (NetWork Analysis) of this package.

Usage

HTSanalyzeR4cellHTS2(		
	normCellHTSobject,
	scoreSign = "-",
	scoreMethod = "zscore",
	summarizeMethod = "mean",
	annotationColumn = "GeneID",
	species = "Dm",
	initialIDs = "FlybaseCG",
	duplicateRemoverMethod = "max",
	orderAbsValue = FALSE,
	listOfGeneSetCollections,
	cutoffHitsEnrichment = 2,
	pValueCutoff = 0.05,
	pAdjustMethod = "BH",
	nPermutations = 1000,
	minGeneSetSize = 15,
	exponent = 1,
	keggGSCs,
	goGSCs,
	nwStatsControls = "neg",
	nwStatsAlternative = "two.sided",
	nwStatsTests = "T-test",
	nwStatsColumns = c("t.test.pvalues.two.samples", "t.test.pvalues.one.sample"),
	nwAnalysisFdr = 0.001,
	nwAnalysisGenetic = FALSE,
	interactionMatrix = NULL,
	nwAnalysisOrder = 2,
	ntop = NULL,
	allSig = TRUE,
	reportDir = "HTSanalyzerReport",
	verbose = TRUE
)

Arguments

normCellHTSobject	a normalized, configured and annotated cellHTS object
scoreSign	a single character value specifying the 'sign' argument for the scoring function from cellHTS2 (see 'scoreReplicates')
scoreMethod	a single character value specifying the 'method' argument for the scoring function from cellHTS2 (see 'scoreReplicates')
summarizeMethod	a summary argument for the summarization function from cellHTS2 (see 'summarizeReplicates')
annotationColumn	a single character value specifying the name of the column in the fData (cellHTSobject) data frame from which the feature identifiers will be extracted
species	a single character value specifying the species for which the data should be read. The current version supports one of the following species: "Dm" ("Drosophila_melanogaster"), "Hs" ("Homo_sapiens"), "Rn" ("Rattus_norvegicus"), "Mm" ("Mus_musculus"), "Ce" ("Caenorhabditis_elegans").
initialIDs	a single character value specifying the type of initial identifiers for input 'geneList'. Current version can take one of the following types: "Ensembl.transcript", "Ensembl.prot", "Ensembl.gene", "Entrez.gene", "RefSeq", "Symbol" and "GenBank" for all supported species; "Flybase", "FlybaseCG" and "FlybaseProt" in addition for Drosophila Melanogaster; "wormbase" in addition for Caenorhabditis Elegans.
duplicateRemoverMethod	a single character value specifying the method to remove the duplicates (should the minimum, maximum or average observation for a same construct be kept). The current version provides "min" (minimum), "max" (maximum), "average" and "fc.avg" (fold change average). The minimum and maximum should be understood in terms of absolute values (i.e. min/max effect, no matter the sign). The fold change average method converts the fold changes to ratios, averages them and converts the average back to a fold change.
orderAbsValue	a single logical value determining whether the values should be converted to absolute value and then ordered (if TRUE), or ordered as they are (if FALSE).
listOfGeneSetCollections	a list of gene set collections (a 'gene set collection' is a list of gene sets). Even if only one collection is being tested,it must be entered as an element of a 1-element list, e.g. ListOfGeneSetCollections = list(YourOneGeneSetCollection). Naming the elements of listOfGeneSetCollections will result in these names being associated with the relevant data frames in the output (meaningful names are advised)
cutoffHitsEnrichment	a single numeric or integer value specifying the cutoff that is used in the definition of the hits for the hypergeometric tests in the over- representation analysis. This cutoff is used in absolute value, since it is applied on scores, i.e. a cutoff of 2 when using z-scores means that we are selecting values that are two standard deviations away from the median of all samples. Therefore, the cutoff should be a positive number.
pValueCutoff	a single numeric value specifying the cutoff for p-values considered significant
pAdjustMethod	a single character value specifying the p-value adjustment method to be used (see 'p.adjust' for details)
nPermutations	a single integer or numeric value specifying the number of permutations for deriving p-values in GSEA
minGeneSetSize	a single integer or numeric value specifying the minimum number of elements in a gene set that must map to elements of the gene universe. Gene sets with fewer than this number are removed from both hypergeometric analysis and GSEA.
exponent	a single integer or numeric value used in weighting phenotypes in GSEA (see "gseaScores" function)
keggGSCs	a character vector of names of all KEGG gene set collections. This will help create web links for KEGG terms.
goGSCs	a character vector of names of all GO gene set collections. This will help create web links for GO terms.
nwStatsControls	a single character value specifying the name of the controls to be used as a control population in the two-sample tests (this HAS to be corresponding to how these control wells have been annotated in the column "controlStatus" of the fData(cellHTSobject) data frame). If nothing is specified, the function will look for negative controls labelled "neg".
nwStatsAlternative	a single character value specifying the alternative hypothesis: "two.sided", "less" or "greater"
nwStatsTests	a single character value specifying the tests to be performed: "T-test", "MannWhitney" or "RankProduct". If nothing is specified, all three tests will be performed. Be aware that the Rank Product test is slower than the other two, and returns a percent false discovery (equivalent to a FDR, not a p-value).
nwStatsColumns	a character vector of any (relevant, i.e. that is produced in the tests) combination of "t.test.pvalues.two.samples", "t.test.pvalues.one.sample", "mannW.test.pvalues.one.sample", "mannW.test.pvalues.two.samples", "rank.product.pfp.greater", "rank.product.pfp.less"
nwAnalysisFdr	a single numeric value specifying the FDR used in the networkAnalysis function for the scores calculation
nwAnalysisGenetic	a single logical value indicating if the genetic interaction data of the Biogrid dataset were kept in the network analysis
interactionMatrix	an interaction matrix including columns 'InteractionType', 'InteractorA' and 'InteractorB'. If this matrix is available, the interactome can be directly built based on it.
nwAnalysisOrder	the order used in the networkAnalysis function for the scores calculation
ntop	the number of plots to be produced for the GSEA analysis. For each gene set collection, plots are produced for the "nplots" most significant p-values.
allSig	a single logical value indicating whether or not to generate plots for all significant gene sets. A gene set is significant if its corresponding adjusted p-value is less than the pValueCutoff set in function analyze (see function analyze for more details).
reportDir	a single character value specifying the directory to store reports
verbose	a single logical value indicating to display detailed messages (when verbose= TRUE) or not (when verbose=FALSE)

Details

This pipeline function performs gene set over-representation, enrichment analyses and identifies enriched subnetworks, writes a report and saves results as 'RData' in the user-specified directory.

For the KEGG and GO gene sets, this function produces links to the relevant database entries. To make the report more readable, the user can append a column named 'Gene.Set.Term' to each data frame in the result slot of the object of class GSCA using the method appendGSTerms. To ensure that the function works properly, the gene set ID should comply with the following requirements:

• for KEGG 3 letters (species ID) followed by a set of numbers and then any number of non-digit characters

• for GO "GO:" immediately followed by the GO term number (digits only) and any number of non-digits characters.

Value

Produce a set of html pages and save GSCA and NWA objects. The report starts from the page called "index.html" to navigate those pages.

Author(s)

Camille Terfve, Xin Wang

Examples

## Not run: 
library(org.Dm.eg.db)
library(GO.db)
library(KEGG.db)
library(cellHTS2)
library(BioNet)
##library(igraph)
#prepare data from cellHTS2
experimentName <- "KcViab"
dataPath <- system.file(experimentName, package = "cellHTS2")
x <- readPlateList("Platelist.txt", name = experimentName, path=dataPath,
verbose=TRUE)
x <- configure(x, descripFile = "Description.txt", confFile = 
"Plateconf.txt",logFile = "Screenlog.txt", path = dataPath)
xn <- normalizePlates(x, scale = "multiplicative", log = FALSE, method =
"median", varianceAdjust = "none")
xn <- annotate(xn, geneIDFile = "GeneIDs_Dm_HFA_1.1.txt",path = dataPath)
cellHTS2DrosoData<-xn
#prepare a list of gene set collections
GO_MF <- GOGeneSets(species = "Dm",ontologies = c("MF"))
PW_KEGG <- KeggGeneSets(species = "Dm")
gscList <- list(GO_MF = GO_MF, PW_KEGG = PW_KEGG)
#note: set computer cluster here to do parallel computing
#options(cluster=makeCluster(4,"SOCK"))
HTSanalyzeR4cellHTS2(
	normCellHTSobject=cellHTS2DrosoData,
	annotationColumn="GeneID",
	species=c("Dm"),
	initialIDs="FlybaseCG",
	listOfGeneSetCollections=gscList,
	cutoffHitsEnrichment=2,
	minGeneSetSize=200,
	keggGSCs="PW_KEGG",
	goGSCs=c("GO_MF"),
	allSig=TRUE,
	reportDir="HTSanalyzerReport",
	verbose=TRUE
)
#note: stop the cluster if you created
#if(is(getOption("cluster"),"cluster")) {
#	stopCluster(getOption("cluster"))
#	options(cluster=NULL)
#}
##browse the index page
browseURL(file.path(getwd(), "HTSanalyzerReport", "index.html"))

## End(Not run)

HTSanalyzeR documentation built on Oct. 31, 2019, 7:10 a.m.