Start of development version, bug fixes and new gene annotation feature

Author: jlmaier12

NEWS.html

@@ -1,10 +1,13 @@
 # ProActive (development version)
 
-# Changes in 0.1.0
+# Changes in 0.1.0.9000
 + Changed name of `ProActive()` to `ProActiveDetect()`
 + Added ability to search read coverage plots for gene annotations of 
 interest with the geneAnnotationSearch() function
 + Fixed bug with pattern-match start and stop positions on genome/contig chunks
++ Bug fixes with contig chunking parameter
+
+# Version 0.1.0 submitted to CRAN!
 
 # Changes in 0.0.2
 + Changes for CRAN re-submission after review

NEWS.md

@@ -0,0 +1,21 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX
+
+AutoAppendNewline: Yes
+StripTrailingWhitespace: Yes
+
+BuildType: Package
+PackageUseDevtools: Yes
+PackageInstallArgs: --no-multiarch --with-keep.source
+PackageRoxygenize: rd,collate,namespace

ProphageActivityTool.Rproj

@@ -75,15 +75,15 @@ ProActiveDetect <- function(pileup, mode, gffTSV, windowSize = 1000, chunkContig
     pileup <- contigChunks(pileup, chunkSize)
   }
   if(verbose){message("Starting pattern-matching...")}
-  patternMatchSummary <- patternMatcher(pileup, windowSize, minSize, maxSize, mode, minContigLength, verbose)
+  patternMatchSummary <- patternMatcher(pileup, windowSize, minSize, maxSize, mode, minContigLength, chunkContigs, verbose)
   if (IncludeNoPatterns) {
     classifList <- patternMatchSummary[[1]]
   } else {
     classifList <- removeNoPatterns(patternMatchSummary[[1]])
   }
   filteredOutContigsDf <- patternMatchSummary[[2]]
   if(verbose){message("Summarizing pattern-matching results")}
-  summaryTable <- classifSumm(pileup, patternMatchSummary[[1]], windowSize, mode, chunkSize)
+  summaryTable <- classifSumm(pileup, patternMatchSummary[[1]], windowSize, mode, chunkContigs, chunkSize)
   if (missing(gffTSV) == FALSE) {
     if(verbose){message("Finding gene predictions in elevated or gapped regions of read coverage...")}
     elevGapSummList <- removeNoPatterns(patternMatchSummary[[1]])
@@ -116,7 +116,7 @@ ProActiveDetect <- function(pileup, mode, gffTSV, windowSize = 1000, chunkContig
   table <- (table(summaryTable[, 2]))
   if(verbose){message(paste0(capture.output(table), collapse = "\n"))}
   if(mode == "genome" || (mode == "metagenome" & chunkContigs == TRUE)){
-    linkChunks(classifList, pileup, windowSize, mode, verbose)
+    linkChunks(classifList, pileup, windowSize, mode, chunkContigs, verbose)
   }
   if (missing(saveFilesTo) == FALSE) {
     ifelse(!dir.exists(paths = paste0(saveFilesTo, "\\ProActiveOutput")),

R/ProActiveDetect.R

@@ -10,9 +10,12 @@
 #' Options are 100, 200, 500, 1000 ONLY. Default is 1000.
 #' @param chunkSize If `mode`="genome" OR if `mode`="metagenome" and `chunkContigs`=TRUE,
 #' chunk the genome or contigs, respectively, into smaller subsets for pattern-matching.
+#' @param chunkContigs TRUE or FALSE, If TRUE and `mode`="metagenome", contigs longer
+#' than the `chunkSize` will be 'chunked' into smaller subsets and pattern-matching
+#' will be performed on each subset. Default is FALSE.
 #' @param mode Either "genome" or "metagenome"
 #' @keywords internal
-classifSumm <- function(pileup, bestMatchList, windowSize, mode, chunkSize) {
+classifSumm <- function(pileup, bestMatchList, windowSize, mode, chunkContigs, chunkSize) {
   if (length(bestMatchList) == 0) {
     stop("No pattern-matches detected")
   }
@@ -34,7 +37,7 @@ classifSumm <- function(pileup, bestMatchList, windowSize, mode, chunkSize) {
   classification <- vapply(seq_along(bestMatchList), function(i) {bestMatchList[[i]][[7]]}, character(1))
   matchSize <- vapply(seq_along(bestMatchList), function(i) {
     pileupSubset <- pileup[which(pileup[, 1] == bestMatchList[[i]][[8]]), ]
-    pileupSubset <- changewindowSize(pileupSubset, windowSize, mode)
+    pileupSubset <- changewindowSize(pileupSubset, windowSize, chunkContigs, mode)
     (length(seq(pileupSubset[bestMatchList[[i]][[4]], 3], pileupSubset[bestMatchList[[i]][[5]], 3], windowSize)) - 1) * windowSize},
     numeric(1))
   classifSummTable <- cbind.data.frame(refName, classification, elevRatio, startPos, endPos, matchSize)

R/bestMatchListFunctions.R

@@ -58,7 +58,7 @@ contigChunks <- function(pileup, chunkSize) {
 #' 100 bp windows/bins.
 #' @param chunkSize If `mode`="genome" OR if `mode`="metagenome" and `chunkContigs`=TRUE,
 #' chunk the genome or contigs, respectively, into smaller subsets for pattern-matching.
-#' `chunkSize` determines the size (in bp) of each 'chunk'. Default is 50000.
+#' `chunkSize` determines the size (in bp) of each 'chunk'. Default is 100000.
 #' @keywords internal
 genomeChunks <- function(pileup, chunkSize) {
   refName <- rep(NA, nrow(pileup))
@@ -99,10 +99,13 @@ genomeChunks <- function(pileup, chunkSize) {
 #' 100 bp windows/bins.
 #' @param windowSize The number of basepairs to average read coverage values over.
 #' @param mode Either "genome" or "metagenome"
+#' @param chunkContigs TRUE or FALSE, If TRUE and `mode`="metagenome", contigs longer
+#' than the `chunkSize` will be 'chunked' into smaller subsets and pattern-matching
+#' will be performed on each subset. Default is FALSE.
 #' @param verbose TRUE or FALSE. Print progress messages to console. Default is TRUE.
 #' @importFrom stringr str_extract
 #' @keywords internal
-linkChunks <- function(bestMatchList, pileup, windowSize, mode, verbose){
+linkChunks <- function(bestMatchList, pileup, windowSize, mode, chunkContigs, verbose){
   potLink <- rep(NA, length(bestMatchList))
   refNames <- vapply(seq_along(bestMatchList), function(i){bestMatchList[[i]][[8]]}, character(1))
   classifVector <- vapply(seq_along(bestMatchList), function(i){
@@ -111,7 +114,7 @@ linkChunks <- function(bestMatchList, pileup, windowSize, mode, verbose){
     startPos <- bestMatchList[[i]][[4]]
     endPos <- bestMatchList[[i]][[5]]
     pileupSubset <- pileup[which(pileup[, 1] == refName), ]
-    pileupSubset <- changewindowSize(pileupSubset, windowSize, mode)
+    pileupSubset <- changewindowSize(pileupSubset, windowSize, chunkContigs, mode)
     if(grepl("chunk", refName) == FALSE) {
       "NoChunk"
     } else if(classification == "NoPattern") {
@@ -140,7 +143,7 @@ linkChunks <- function(bestMatchList, pileup, windowSize, mode, verbose){
         classifDf[rightIdx,3] <- "no"
       }
     }
-    if(verbose == TRUE){
+    if(verbose){
       if("link" %in% classifDf[,3]){
       linkIdxs <- which(classifDf[,3] == "link")
       lapply(seq_along(linkIdxs), function(x){

R/chunkingFunctions.R

@@ -14,10 +14,13 @@
 #' @param mode Either "genome" or "metagenome".
 #' @param minContigLength The minimum contig/chunk size (in bp) to perform pattern-matching
 #' on. Default is 25000.
+#' @param chunkContigs TRUE or FALSE, If TRUE and `mode`="metagenome", contigs longer
+#' than the `chunkSize` will be 'chunked' into smaller subsets and pattern-matching
+#' will be performed on each subset. Default is FALSE.
 #' @param verbose TRUE or FALSE. Print progress messages to console. Default is TRUE.
 #' @importFrom stats na.omit
 #' @keywords internal
-patternMatcher <- function(pileup, windowSize, minSize, maxSize, mode, minContigLength, verbose) {
+patternMatcher <- function(pileup, windowSize, minSize, maxSize, mode, minContigLength, chunkContigs, verbose) {
   refNames <- unique(pileup[, 1])
   bestMatchList <- vector(mode='list', length=length(refNames))
   filteredOutContigs <- rep(NA, length(refNames))
@@ -51,7 +54,7 @@ patternMatcher <- function(pileup, windowSize, minSize, maxSize, mode, minContig
       C <- C + 1
       next
     }
-    pileupSubset <- changewindowSize(pileupSubset, windowSize, mode)
+    pileupSubset <- changewindowSize(pileupSubset, windowSize, chunkContigs, mode)
     noPatternBestMatch <- noPattern(pileupSubset)
     partialElevBestMatch <- partialElevGap(pileupSubset, windowSize, minSize, maxSize)
     fullElevBestMatch <- fullElevGap(pileupSubset, windowSize, minSize, maxSize, "Elevation")

R/patternMatcher.R

@@ -42,7 +42,7 @@ plotProActiveResults <- function(pileup, ProActiveResults, elevFilter, saveFiles
       return(NULL)
     }
     pileupSubset <- pileup[which(pileup[, 1] == refName), ]
-    pileupSubset <- changewindowSize(pileupSubset, windowSize, mode)
+    pileupSubset <- changewindowSize(pileupSubset, windowSize, contigChunk, mode)
     patternMatch <- patternBuilder(pileupSubset, patternMatches[[i]])
     matchLength <- matchInfo[, 6]
     plot <- ggplot(data = patternMatch, aes(x = position, y = coverage)) +

R/plotProActiveResults.R

@@ -6,9 +6,12 @@
 #'  currently being assessed.
 #' @param windowSize The number of basepairs to average read coverage values over.
 #' Options are 100, 200, 500, 1000 ONLY. Default is 1000.
+#' @param chunkContigs TRUE or FALSE, If TRUE and `mode`="metagenome", contigs longer
+#' than the `chunkSize` will be 'chunked' into smaller subsets and pattern-matching
+#' will be performed on each subset. Default is FALSE.
 #' @param mode Either "genome" or "metagenome"
 #' @keywords internal
-changewindowSize <- function(pileupSubset, windowSize, mode) {
+changewindowSize <- function(pileupSubset, windowSize, chunkContigs, mode) {
   coverage <- vector()
   X <- 0
   Y <- windowSize / 100
@@ -18,7 +21,7 @@ changewindowSize <- function(pileupSubset, windowSize, mode) {
     Y <- Y + (windowSize / 100)
     if (Y > nrow(pileupSubset)) break
   }
-  if (mode == "genome") {
+  if (mode == "genome" || chunkContigs == TRUE) {
     position <- seq(pileupSubset[1, 3], pileupSubset[nrow(pileupSubset), 3], length.out = length(coverage))
   } else {
     position <- seq(windowSize, length(coverage) * windowSize, windowSize)

R/reformatInputFunctions.R

@@ -0,0 +1,24 @@
+## Resubmission
+This is a resubmission. In this version (0.1.0) I have:
+
+* Changed name of `ProActive()` to `ProActiveDetect()`
+
+* Added geneAnnotationSearch() to main functions available to users
+
+* Fixed bug with pattern-match start and stop positions on genome/contig chunks
+
+## Resubmission
+This is a resubmission. In this version (0.0.2) I have:
+
+* Changed description text so it does not start with 'this package does..'
+
+* Removed all use of directed quotation marks in description text
+
+* Fixed a bug where use of the elevFilter parameter in the plotProActiveResults 
+function produced an error message
+
+## R CMD check results
+
+0 errors | 0 warnings | 1 note
+
+* This is a new release.