R/computeIC.R

In GOSemSim: GO-terms Semantic Similarity Measures

##' @importFrom AnnotationDbi as.list
##' @importFrom GO.db GOBPOFFSPRING
##' @importFrom GO.db GOCCOFFSPRING
##' @importFrom GO.db GOMFOFFSPRING
computeIC <- function(goAnno, ont) {
    ## goAnno, see godata function
    if (!exists(".GOSemSimEnv")) .initial()
    .GOSemSimEnv <- get(".GOSemSimEnv", envir=.GlobalEnv)
    godata <- get("gotbl", envir=.GOSemSimEnv)        
    
    goids <- unique(godata[godata$Ontology == ont, "go_id"])
    ## all GO terms appearing in an given ontology ###########
    goterms=goAnno$GO
    gocount <- table(goterms)
    ## goid of specific organism and selected category.
    goname  <- names(gocount) 

    ## ensure goterms not appearing in the specific annotation have 0 frequency..
    go.diff        <- setdiff(goids, goname)
    m              <- double(length(go.diff))
    names(m)       <- go.diff
    gocount        <- as.vector(gocount)
    names(gocount) <- goname
    gocount        <- c(gocount, m)

    Offsprings <- switch(ont,
                         MF = AnnotationDbi::as.list(GOMFOFFSPRING),
                         BP = AnnotationDbi::as.list(GOBPOFFSPRING),
                         CC = AnnotationDbi::as.list(GOCCOFFSPRING))
        
    cnt <- gocount[goids] + sapply(goids, function(i) sum(gocount[Offsprings[[i]]], na.rm=TRUE))
    names(cnt) <- goids
    
    ## the probabilities of occurrence of GO terms in a specific corpus.
    p <- cnt/sum(gocount)
    ## IC of GO terms was quantified as the negative log likelihood.
    IC <- -log(p)
    return(IC)
}