R/clusters_cor.R

In ViSEAGO: ViSEAGO: a Bioconductor package for clustering biological functions using Gene Ontology and semantic similarity

#' @title Compute distance matrix between dendrograms partitions.
#' @description Build a distance or correlation matrix between partitions from dendrograms.
#' @importFrom data.table data.table setorderv
#' @importFrom igraph compare
#' @family GO_clusters
#' @param clusters  a \code{list} of \code{\link{GO_clusters-class}} objects,
#' from \code{\link{GOterms_heatmap}} or \code{\link{GOclusters_heatmap}}, named as \code{character}.
#' @param method available methods  ("vi", "nmi", "split.join", "rand", or "adjusted.rand") from \pkg{igraph} package \code{\link[igraph]{compare}} function.
#' @return a distance or a correlation matrix.
#' @references
#' Csardi G, Nepusz T: The igraph software package for complex network research, InterJournal, Complex Systems 1695. 2006. http://igraph.org.
#' @include GO_clusters.R
#' @examples
#' # load example object
#' data(
#'     myGOs,
#'     package="ViSEAGO"
#' )
#'
#' \dontrun{
#' # compute Semantic Similarity (SS)
#' myGOs<-ViSEAGO::compute_SS_distances(
#'     myGOs,
#'     distance=c("Resnik","Lin","Rel","Jiang","Wang")
#' )
#'
#' # Resnik distance GO terms heatmap
#' Resnik_clusters_wardD2<-ViSEAGO::GOterms_heatmap(
#'     myGOs,
#'     showIC=TRUE,
#'     showGOlabels=TRUE,
#'     GO.tree=list(
#'         tree=list(
#'             distance="Resnik",
#'             aggreg.method="ward.D2"
#'         ),
#'         cut=list(
#'             dynamic=list(
#'                 deepSplit=2,
#'                 minClusterSize =2
#'             )
#'         )
#'     ),
#'     samples.tree=NULL
#' )
#'
#' # Lin distance GO terms heatmap
#' Lin_clusters_wardD2<-ViSEAGO::GOterms_heatmap(
#'     myGOs,
#'     showIC=TRUE,
#'     showGOlabels=TRUE,
#'     GO.tree=list(
#'         tree=list(
#'             distance="Lin",
#'             aggreg.method="ward.D2"
#'         ),
#'         cut=list(
#'             dynamic=list(
#'                 deepSplit=2,
#'                 minClusterSize =2
#'             )
#'         )
#'     ),
#'     samples.tree=NULL
#' )
#'
#' # Resnik distance GO terms heatmap
#' Rel_clusters_wardD2<-ViSEAGO::GOterms_heatmap(
#'     myGOs,
#'     showIC=TRUE,
#'     showGOlabels=TRUE,
#'     GO.tree=list(
#'         tree=list(
#'             distance="Rel",
#'             aggreg.method="ward.D2"
#'         ),
#'         cut=list(
#'             dynamic=list(
#'                 deepSplit=2,
#'                 minClusterSize =2
#'             )
#'         )
#'     ),
#'     samples.tree=NULL
#' )
#'
#' # Resnik distance GO terms heatmap
#' Jiang_clusters_wardD2<-ViSEAGO::GOterms_heatmap(
#'     myGOs,
#'     showIC=TRUE,
#'     showGOlabels=TRUE,
#'     GO.tree=list(
#'         tree=list(
#'             distance="Jiang",
#'             aggreg.method="ward.D2"
#'         ),
#'         cut=list(
#'             dynamic=list(
#'                 deepSplit=2,
#'                 minClusterSize =2
#'             )
#'         )
#'     ),
#'     samples.tree=NULL
#' )
#'
#' # Resnik distance GO terms heatmap
#' Wang_clusters_wardD2<-ViSEAGO::GOterms_heatmap(
#'     myGOs,
#'     showIC=TRUE,
#'     showGOlabels=TRUE,
#'     GO.tree=list(
#'         tree=list(
#'             distance="Wang",
#'             aggreg.method="ward.D2"
#'         ),
#'         cut=list(
#'             dynamic=list(
#'                 deepSplit=2,
#'                 minClusterSize =2
#'             )
#'         )
#'     ),
#'     samples.tree=NULL
#' )
#' }
#' # clusters to compare
#' clusters<-list(
#'     Resnik="Resnik_clusters_wardD2",
#'     Lin="Lin_clusters_wardD2",
#'     Rel="Rel_clusters_wardD2",
#'     Jiang="Jiang_clusters_wardD2",
#'     Wang="Wang_clusters_wardD2"
#' )
#'
#' \dontrun{
#' # global dendrogram clustering correlation
#' clust_cor<-ViSEAGO::clusters_cor(
#'     clusters,
#'     method="adjusted.rand"
#' )
#' }
#' @name clusters_cor
#' @rdname clusters_cor-methods
#' @exportMethod clusters_cor
setGeneric(
    name="clusters_cor",
    def=function(clusters,method="adjusted.rand"){
        standardGeneric("clusters_cor")
    }
)

#' @rdname clusters_cor-methods
#' @aliases clusters_cor
setMethod(
    "clusters_cor",
    signature(
        clusters="list",
        method="character"
    ),
    definition=function(clusters,method){

        # method
        method=match.arg(
            method,c("vi", "nmi", "split.join", "rand","adjusted.rand")
        )

        ## extract clusters and GO.ID

        # extract clusters and GO.ID
        Clusters<-lapply(clusters,function(x){

            # load the target object
            object=get(x[[1]])

            # check class
            if(!is(object,"GO_clusters")){
                stop("object must come from ViSEAGO::GOterms_heatmap()")
            }

            # extract GO.ID and correspond cluster assignation to a list
            object<- slot(slot(object,"enrich_GOs"),"data")[,c("GO.ID","GO.cluster"),with=FALSE]

            # ordering by GO.ID
            setorderv(object,"GO.ID")

            # return the object
            object$GO.cluster
        })

        # add names
        names(Clusters)<-names(clusters)

        ## extract clusters and GO.ID

        # number of list elements
        n_list <- length(Clusters)

        # build o matrix of one
        the_cor <- matrix(1, n_list, n_list)

        # build pairwise combinations
        pairwise_combn <- combn(n_list, 2)

        # for each combination
        for (i in seq_len(ncol(pairwise_combn))) {

            # extract l1 elements
            l1 <- pairwise_combn[1, i]

            # extract l2 elements
            l2 <- pairwise_combn[2, i]

            # add the value to the_cor
            the_cor[l1, l2] <- the_cor[l2, l1] <-compare(Clusters[[l1]], Clusters[[l2]],method=method)
        }

        # add headers
        rownames(the_cor) <- colnames(the_cor)<-names(Clusters)

        # return the results
        the_cor
    }
)