diff --git a/DESCRIPTION b/DESCRIPTION
index dbb17b0fe90fbb7a8792091335515bc0e7c5a3db..9c57a6efb1f14649c47830183634721135b0ae1d 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,5 +1,5 @@
Package: ViSEAGO
-Version: 1.3.7
+Version: 1.3.8
Title: ViSEAGO: a Bioconductor package for clustering biological functions using Gene Ontology and semantic similarity
Author: Aurelien Brionne [aut, cre],
Amelie Juanchich [aut],
@@ -51,7 +51,7 @@ URL: https://www.bioconductor.org/packages/release/bioc/html/ViSEAGO.html,
https://forgemia.inra.fr/UMR-BOA/ViSEAGO
BugReports: https://forgemia.inra.fr/UMR-BOA/ViSEAGO/issues
biocViews: Software, Annotation, GO, GeneSetEnrichment, MultipleComparison, Clustering, Visualization
-RoxygenNote: 7.1.0
+RoxygenNote: 7.1.1
Collate:
'genomic_ressource.R'
'Bioconductor2GO.R'
diff --git a/R/enrich_GO_terms.R b/R/enrich_GO_terms.R
index 8fa5d88426127c4480ebf0f8ac5a4f7f882dada7..97c97c506b54e36ed1b81a6ad36124a76282d4b1 100644
--- a/R/enrich_GO_terms.R
+++ b/R/enrich_GO_terms.R
@@ -14,6 +14,7 @@ setClass(
same_genes_background="logical",
input="list",
ont="character",
+ cutoff="list",
topGO="list",
data="data.table"
)
@@ -33,20 +34,7 @@ setMethod(
Data<-Data[,grep("\\.pvalue",names(Data)),with=FALSE]
# pvalues threshlod according condition
- p<-vapply(slot(object,"topGO"),function(x){
-
- # unlist
- x=unlist(x)
-
- # extract pvalue threshold
- as.numeric(
- sub(
- "^.+<",
- "",
- x[grep("test_name",names(x))]
- )
- )
- },0)
+ p<-unlist(slot(object,"cutoff"))
# count significant pvalues by condition
Data<-lapply(seq_len(ncol(Data)),function(x){
diff --git a/R/merge_enrich_terms.R b/R/merge_enrich_terms.R
index 0c06c275d9e43c382d0acb77e42a37a9b4b15f6c..17c482bcdd82d9331a171362d73436f692442362 100644
--- a/R/merge_enrich_terms.R
+++ b/R/merge_enrich_terms.R
@@ -9,13 +9,14 @@
#' @family GO_terms
#' @param Input a list containing named elements. Each element must contain the name of \code{\link[topGO]{topGOdata-class}}
#' object created by \code{\link{create_topGOdata}} method and the associated \code{\link[topGO]{topGOresult-class}} object(s).
+#' @param cutoff default pvalue cutoff (default to 0.01). Several cutoff can be use in the same order as list elements.
#' @param envir objects environment (default to .GlobalEnv).
#' @details This method extracts for each result of GO enrichment test (\code{\link[topGO]{topGOresult-class}} object) and
#' corresponding GO annotations (\code{\link[topGO]{topGOdata-class}} object):
#' informations about GO term (identifiant, name, and description),
#' gene frequency (number of significant genes / Annotated genes), pvalue, -log10(pvalue), significant genes
#' identifiants (GeneID, or Ensembl ID, or uniprot accession), and gene symbols.
-#' At the last, this method builds a merged data.table of enriched GO terms (p<0.01)
+#' At the last, this method builds a merged data.table of enriched GO terms
#' at least once and provides all mentionned columns.
#' @return an \code{\link{enrich_GO_terms-class}} object.
#' @references
@@ -129,6 +130,7 @@ setGeneric(
name="merge_enrich_terms",
def=function(
Input,
+ cutoff=0.01,
envir=.GlobalEnv
){
standardGeneric("merge_enrich_terms")
@@ -142,7 +144,19 @@ setMethod(
signature(
Input="list"
),
- definition=function(Input,envir){
+ definition=function(Input,cutoff,envir){
+
+ ## cutoff
+
+ # check cutoff length according Input
+ if(length(cutoff)>1 & length(cutoff)!=length(Input)){
+ stop("cutoff must be a single value to repet for each list element, or the same length than list")
+ }
+
+ # repet cutoff value according Input length
+ if(length(cutoff)==1){
+ cutoff<-rep(cutoff,length(Input))
+ }
## check ontology
@@ -202,6 +216,9 @@ setMethod(
# build topGO summary
topGO_summary=lapply(seq_along(Input),function(x){
+ # keep pos
+ pos=x
+
# extract quering objects names
x=Input[[x]]
@@ -266,7 +283,7 @@ setMethod(
GO_scored=length(slot(x[[y]],"score")),
# significant GOs according cutOff
- GO_significant=table(slot(x[[y]],"score")<as.numeric(sub("^.+[[:space:]]","",slot(x[[y]],"testName"))))[2],
+ GO_significant=table(slot(x[[y]],"score")<cutoff[pos])[2],
# feasibles genes
feasible_genes=slot(x[[y]],"geneData")[1],
@@ -329,7 +346,7 @@ setMethod(
# extract names of enrich terms
as.vector(
names(
- pvalues[pvalues<as.numeric(sub("^.+[[:space:]]","",slot(Data[[pos]],"testName")))]
+ pvalues[pvalues<cutoff[x]]
)
)
})
@@ -595,12 +612,23 @@ setMethod(
# if db match to Ensembl
if(db[1]=="Ensembl"){
- # connect to Ensembl
- mart<-useEnsembl(
- "ensembl",
- host=db[2],
- version=db[6]
- )
+ if(db[3]!="http://grch37.ensembl.org"){
+
+ # connect to Ensembl
+ mart<-useEnsembl(
+ biomart="genes",
+ host=db[3],
+ version=db[6]
+ )
+
+ }else{
+ # connect to Ensembl
+ mart<-useEnsembl(
+ biomart="genes",
+ GRCh =37,
+ version=db[6]
+ )
+ }
# connect to ensembl specified dataset
myspecies<-useDataset(
@@ -720,11 +748,8 @@ setMethod(
# remove NA in GO.Id column
Results<-Results[!is.na(Results$GO.ID)]
- # extract pvalue threshold
- p<-as.numeric(sub("^.+[[:space:]]","",slot(algorithm[[1]],"testName")))
-
# Remove gene ID and symbol if GO term not significant
- Results[Results$pvalue>=p,`:=`(Significant_genes=NA,Significant_genes_symbol=NA)]
+ Results[Results$pvalue>=cutoff[x],`:=`(Significant_genes=NA,Significant_genes_symbol=NA)]
if(!is.null(names(Input))){
@@ -792,6 +817,7 @@ setMethod(
input=input,
ont=check.onto,
topGO=topGO_summary,
+ cutoff=list(cutoff),
data=allResults
)
}
diff --git a/README.md b/README.md
index 6c2ce311424e9ef4dda85af7aad1e074d5b7c8c7..8ad7a84e0259b055e947b9e1367b014fc0aa5990 100644
--- a/README.md
+++ b/README.md
@@ -27,7 +27,7 @@ remotes::install_gitlab(
devtools::build("ViSEAGO")
# install package (from R console)
- install.packages("ViSEAGO_1.3.7.tar.gz", repos = NULL, type = "source")
+ install.packages("ViSEAGO_1.3.8.tar.gz", repos = NULL, type = "source")
```
## Citation
diff --git a/man/merge_enrich_terms-methods.Rd b/man/merge_enrich_terms-methods.Rd
index 8145dca8bfc1e32dec03f3df0669034db5766b90..5fab93c95e8830152beb87f02b8f73f5b9c8b0c4 100644
--- a/man/merge_enrich_terms-methods.Rd
+++ b/man/merge_enrich_terms-methods.Rd
@@ -5,14 +5,16 @@
\alias{merge_enrich_terms,list-method}
\title{Merge enriched GO terms.}
\usage{
-merge_enrich_terms(Input, envir = .GlobalEnv)
+merge_enrich_terms(Input, cutoff = 0.01, envir = .GlobalEnv)
-\S4method{merge_enrich_terms}{list}(Input, envir = .GlobalEnv)
+\S4method{merge_enrich_terms}{list}(Input, cutoff = 0.01, envir = .GlobalEnv)
}
\arguments{
\item{Input}{a list containing named elements. Each element must contain the name of \code{\link[topGO]{topGOdata-class}}
object created by \code{\link{create_topGOdata}} method and the associated \code{\link[topGO]{topGOresult-class}} object(s).}
+\item{cutoff}{default pvalue cutoff (default to 0.01). Several cutoff can be use in the same order as list elements.}
+
\item{envir}{objects environment (default to .GlobalEnv).}
}
\value{
@@ -28,7 +30,7 @@ corresponding GO annotations (\code{\link[topGO]{topGOdata-class}} object):
informations about GO term (identifiant, name, and description),
gene frequency (number of significant genes / Annotated genes), pvalue, -log10(pvalue), significant genes
identifiants (GeneID, or Ensembl ID, or uniprot accession), and gene symbols.
-At the last, this method builds a merged data.table of enriched GO terms (p<0.01)
+At the last, this method builds a merged data.table of enriched GO terms
at least once and provides all mentionned columns.
}
\examples{
diff --git a/vignettes/mouse_bionconductor.Rmd b/vignettes/mouse_bionconductor.Rmd
index f72292131fe6614183adde2e2c9b34de3d35afde..9201dadc9c9f1ff7f3bc8e1b88414160ba4ce99b 100644
--- a/vignettes/mouse_bionconductor.Rmd
+++ b/vignettes/mouse_bionconductor.Rmd
@@ -210,6 +210,7 @@ The printed table contains for each enriched GO terms, additional columns includ
```{r Enrichment_merge}
# merge topGO results
BP_sResults<-ViSEAGO::merge_enrich_terms(
+ cutoff=0.01,
Input=list(
PregnantvsLactate=c(
"BP_PregnantvsLactate",