change the calculation of functional spectrua to Rcpp

Author: gaofeng21cn

DESCRIPTION

@@ -9,7 +9,8 @@ Description: DeepCC is versatile and robust cancer subtyping framework, based on
 License: MIT
 LazyData: TRUE
 RoxygenNote: 5.0.1
-Imports: h2o, parallel, doParallel, foreach
+Imports: h2o, parallel, doParallel, foreach,
+    Rcpp
 URL: http://github.com/CityUHK-CompBio/DeepCC
 BugReports: http://github.com/CityUHK-CompBio/DeepCC/issues
-
+LinkingTo: Rcpp

NAMESPACE

@@ -1,11 +1,14 @@
-# Generated by roxygen2 (4.1.1): do not edit by hand
+# Generated by roxygen2: do not edit by hand
 
 export(calcEnrichmentScore)
 export(cross_validataion)
 export(getDeepCCFeatures)
 export(getDeepCCLabels)
 export(getFunctionalSpectra)
 export(getFunctionalSpectrum)
+export(get_gene_sets)
 export(trainDeepCCModel)
+import(Rcpp)
 importFrom(foreach,"%dopar%")
 importFrom(foreach,foreach)
+useDynLib(DeepCC)

R/FunctionalSpectra.R

@@ -19,33 +19,14 @@ preprocessGeneList <- function(geneList) {
 #' @param geneList a named vector containing the values of gene expression
 #' @param geneSet a vector containing genes to represent a gene set
 #' @return a numeric indicating enrichment score
+#' @useDynLib DeepCC
+#' @import Rcpp
 #' @export
 #' @examples
 #' calcEnrichmentScore(geneList, geneSet)
 calcEnrichmentScore <- function (geneList, geneSet)
 {
-  hits <- (names(geneList) %in% geneSet)
-  nh <- sum(hits)
-  N <- length(geneList)
-  ES <- 0
-  runningES <- rep(0, N)
-  if (nh > N) {
-    stop("Gene Set is larger than Gene List")
-  }
-  else {
-    if (nh) {
-      tmp <- rep(0, N)
-      NR = sum(abs(geneList[hits]))
-      tmp[hits] <- abs(geneList[hits])/NR
-      tmp[!hits] <- -1/(N - nh)
-      runningES <- cumsum(tmp)
-
-      ESmax <- max(runningES)
-      ESmin <- min(runningES)
-      ES <- ifelse(abs(ESmin) > abs(ESmax), ESmin, ESmax)
-    }
-  }
-  ES
+  calcEnrichmentScoreCPP((names(geneList) %in% geneSet), geneList, 1)
 }
 
 #' Generate Functional Spectra

R/RcppExports.R

@@ -0,0 +1,7 @@
+# Generated by using Rcpp::compileAttributes() -> do not edit by hand
+# Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393
+
+calcEnrichmentScoreCPP <- function(Set, Eso, exponent) {
+    .Call('DeepCC_calcEnrichmentScoreCPP', PACKAGE = 'DeepCC', Set, Eso, exponent)
+}
+

R/utils.R

@@ -28,3 +28,18 @@ cross_validataion <- function(fs, labels, fold = 5) {
   error_rate <- 1-mean(as.character(testData.labels) == as.character(pred.lables))
   error_rate
 }
+
+#' Get Gene Sets
+#'
+#' This function extract a list of gene sets from gmt file.
+#'
+#' @param file filename of the gmt file
+#' @return a list containing gene sets by EntrezID
+#' @export
+#' @examples
+#' msigdbv51 <- get_gene_sets("msigdb.v5.1.entrez.gmt")
+
+get_gene_sets <- function(file) {
+  msig <- GSEABase::getGmt(file, geneIdType=EntrezIdentifier())
+  MSigDBv5.1 <- GSEABase::geneIds(msig)
+}