@@ -26,6 +26,7 @@ export(plotAbs)

 export(plotBestNormal)

 export(poolCoverage)

 export(predictSomatic)

+export(preprocessIntervals)

 export(readCoverageFile)

 export(readCurationFile)

 export(runAbsoluteCN)

@@ -16,6 +16,7 @@ API CHANGES

       purity/ploidy solution)

     o changed default of runAbsoluteCN max.logr.sdev to 0.6 (from 0.75)

     o createTargetWeights does not require tumor coverages anymore

+    o calculateBamCoverageByInterval was renamed to preprocessIntervals

@@ -18,7 +18,7 @@

 #' @param keep.duplicates Keep or remove duplicated reads.

 #' @return Returns total and average coverage by intervals.

 #' @author Markus Riester

-#' @seealso \code{\link{calculateGCContentByInterval}

+#' @seealso \code{\link{preprocessIntervals}

 #' \link{correctCoverageBias} \link{runAbsoluteCN}}

 #' @examples

 #'

@@ -90,9 +90,8 @@ log.ratio.cutoffs = c(-0.9, 0.9), failed = NULL, all.genes = FALSE) {

 #' @param gc.gene.file A mapping file that assigns GC content and gene symbols

 #' to each exon in the coverage files. Used for generating gene-level calls.

 #' First column in format CHR:START-END. Second column GC content (0 to 1).

-#' Third column gene symbol. This file can be generated with the \sQuote{GATK

-#' GCContentByInterval} tool or with the

-#' \code{\link{calculateGCContentByInterval}} function.

+#' Third column gene symbol. This file is generated with the 

+#' \code{\link{preprocessIntervals}} function.

 #' @param fun.focal Function for identifying focal amplifications. Defaults to

 #' \code{\link{findFocal}}.

 #' @param args.focal Arguments for focal amplification function.

@@ -13,9 +13,8 @@ globalVariables(names=c("..level.."))

 #' @param gc.gene.file File providing GC content for each exon in the coverage

 #' files. First column in format CHR:START-END. Second column GC content (0 to

 #' 1).  Third column provides gene symbols, which are optional, but used in

-#' \code{\link{runAbsoluteCN}} to generate gene level calls. This file can be

-#' generated with GATK GCContentByInterval tool or with the

-#' \code{\link{calculateGCContentByInterval}} function.

+#' \code{\link{runAbsoluteCN}} to generate gene level calls. This file is 

+#' generated with the \code{\link{preprocessIntervals}} function.

 #' @param output.file Optionally, write file with GC corrected coverage. Can be

 #' read with the \code{\link{readCoverageFile}} function.

 #' @param plot.gc.bias Optionally, plot GC profiles of the pre-normalized and

@@ -27,7 +26,7 @@ globalVariables(names=c("..level.."))

 #' estimation is applied. If the \code{plot.gc.bias} parameter is set as

 #' \code{FALSE}, this will be ignored.

 #' @author Angad Singh, Markus Riester

-#' @seealso \code{\link{calculateGCContentByInterval}}

+#' @seealso \code{\link{preprocessIntervals}}

 #' @examples

 #' 

 #' normal.coverage.file <- system.file("extdata", "example_normal.txt", 

similarity index 88%

rename from R/calculateGCContentByInterval.R

rename to R/preprocessIntervals.R

@@ -1,9 +1,9 @@

-#' Calculates GC content by interval

+#' Preprocess intervals

 #' 

-#' Uses \code{scanFa} from the Rsamtools package to retrieve GC content of

-#' intervals in a reference FASTA file. Can optimize intervals for copy

-#' number calling by tiling long intervals and by including off-target regions.

-#' This optimization largely follows CNVkit. 

+#' Optimize intervals for copy number calling by tiling long intervals and by 

+#' including off-target regions. Uses \code{scanFa} from the Rsamtools package 

+#' to retrieve GC content of intervals in a reference FASTA file. If provided,

+#' will annotate intervals with mappability and replication timing scores.

 #' 

 #' @param interval.file File specifying the intervals. Interval is expected in

 #' first column in format CHR:START-END.  Instead of a file, a \code{GRanges}

@@ -44,25 +44,30 @@

 #'     package="PureCN", mustWork = TRUE)

 #' bed.file <- system.file("extdata", "ex2_intervals.bed", 

 #'     package="PureCN", mustWork = TRUE)

-#' calculateGCContentByInterval(interval.file, reference.file, 

+#' preprocessIntervals(interval.file, reference.file, 

 #'     output.file="gc_file.txt")

 #' 

 #' intervals <- import(bed.file)

-#' calculateGCContentByInterval(intervals, reference.file, 

+#' preprocessIntervals(intervals, reference.file, 

 #'     output.file="gc_file.txt")

 #' 

-#' @export calculateGCContentByInterval

+#' @export preprocessIntervals

 #' @importFrom rtracklayer import

 #' @importFrom Biostrings letterFrequency

 #' @importFrom BiocGenerics unstrand

 #' @importFrom stats aggregate

 #' @importFrom S4Vectors mcols

 #' @importFrom GenomeInfoDb seqlevelsInUse seqlengths seqlevels<-

-calculateGCContentByInterval <- function(interval.file, reference.file,

-output.file = NULL, off.target=FALSE, average.target.width=400, 

-min.off.target.width=20000, average.off.target.width=200000,  

-off.target.padding=-500, mappability=NULL, min.mappability=c(0.5,0.1,0.7),

-reptiming=NULL, off.target.seqlevels=c("targeted", "all")) {

+preprocessIntervals <- function(interval.file, reference.file,

+                                output.file = NULL, off.target = FALSE,

+                                average.target.width = 400,

+                                min.off.target.width = 20000,

+                                average.off.target.width = 200000,

+                                off.target.padding = -500, mappability = NULL,

+                                min.mappability = c(0.5, 0.1, 0.7), 

+                                reptiming = NULL,

+                                off.target.seqlevels=c("targeted", "all")) {

+

     if (class(interval.file)=="GRanges") {

         interval.gr <- .checkIntervals(interval.file)

     } else {

@@ -156,6 +161,18 @@ reptiming=NULL, off.target.seqlevels=c("targeted", "all")) {

     invisible(interval.gr)

 }

+#' Calculates GC content by interval

+#'

+#' This function was renamed to \code{\link{preprocessIntervals}}.

+#'

+#' @param ... Arguments passed to \code{\link{preprocessIntervals}}.

+#' 

+#' @export calculateGCContentByInterval

+calculateGCContentByInterval <- function(...) {

+    .Deprecated("preprocessIntervals")

+    preprocessIntervals(...)

+}    

+

 # this function removes short chromosomes that have no probes (mainly a

 # general way to remove chrM)

 .dropShortUntargetedSeqLevels <- function(offRegions, interval.gr, minSize) {

@@ -161,9 +161,8 @@

 #' @param gc.gene.file A mapping file that assigns GC content and gene symbols

 #' to each exon in the coverage files. Used for generating gene-level calls.

 #' First column in format CHR:START-END. Second column GC content (0 to 1).

-#' Third column gene symbol. This file can be generated with the \sQuote{GATK

-#' GCContentByInterval} tool or with the

-#' \code{\link{calculateGCContentByInterval}} function.

+#' Third column gene symbol. This file is generated with the

+#' \code{\link{preprocessIntervals}} function.

 #' @param max.dropout Measures GC bias as ratio of coverage in AT-rich (GC <

 #' 0.5) versus GC-rich on-target regions (GC >= 0.5). High drop-out might 

 #' indicate that  data was not GC-normalized or that the sample quality might 

@@ -10,16 +10,16 @@ option_list <- list(

         default = NULL,

         help = "Infile specifying target (baits) intervals. Needs to be parsable by rtracklayer."),

     make_option(c("--offtarget"), action = "store_true",

-        default = formals(PureCN::calculateGCContentByInterval)$off.target, 

+        default = formals(PureCN::preprocessIntervals)$off.target, 

         help = "Include off-target regions [default %default]"),

     make_option(c("--targetwidth"), action = "store", type = "integer",

-        default=formals(PureCN::calculateGCContentByInterval)$average.target.width, 

+        default=formals(PureCN::preprocessIntervals)$average.target.width, 

         help = "Split large targets to approximately that size [default %default]"),

     make_option(c("--offtargetwidth"), action = "store", type = "integer",

-        default = formals(PureCN::calculateGCContentByInterval)$average.off.target.width, 

+        default = formals(PureCN::preprocessIntervals)$average.off.target.width, 

         help = "Bin off-target regions to approximately that size [default %default]"),

     make_option(c("--offtargetseqlevels"), action = "store", type = "character",

-        default=formals(PureCN::calculateGCContentByInterval)$off.target.seqlevels[[2]], 

+        default=formals(PureCN::preprocessIntervals)$off.target.seqlevels[[2]], 

         help = "Controls how to deal with chromosomes/contigs not found in infile. One of targeted, all [default %default]"),

     make_option(c("--mappability"), action = "store", type = "character", 

         help = "File parsable by rtracklayer specifying mappability scores of genomic regions."),

@@ -110,7 +110,7 @@ if (!opt$offtarget) {

      " Amplicon data. Add --offtarget to include them.")

 }

-outGC <- calculateGCContentByInterval(intervals, reference.file, 

+outGC <- preprocessIntervals(intervals, reference.file, 

     output.file = outfile, off.target = opt$offtarget, 

     mappability = mappability, average.off.target.width = opt$offtargetwidth,

     reptiming = reptiming, off.target.seqlevels = opt$offtargetseqlevels,

@@ -11,6 +11,6 @@

   The following functions are deprecated and will be made defunct; use

   the replacement indicated below:

   \itemize{

-    \item{none}

+    \item{calculateGCContentByInterval: \code{\link{preprocessIntervals}}}

   }

 }

@@ -47,7 +47,7 @@ coverage <- calculateBamCoverageByInterval(bam.file = bam.file,

 }

 \seealso{

-\code{\link{calculateGCContentByInterval}

+\code{\link{preprocessIntervals}

 \link{correctCoverageBias} \link{runAbsoluteCN}}

 }

 \author{

@@ -1,83 +1,14 @@

 % Generated by roxygen2: do not edit by hand

-% Please edit documentation in R/calculateGCContentByInterval.R

+% Please edit documentation in R/preprocessIntervals.R

 \name{calculateGCContentByInterval}

 \alias{calculateGCContentByInterval}

 \title{Calculates GC content by interval}

 \usage{

-calculateGCContentByInterval(interval.file, reference.file,

-  output.file = NULL, off.target = FALSE, average.target.width = 400,

-  min.off.target.width = 20000, average.off.target.width = 2e+05,

-  off.target.padding = -500, mappability = NULL, min.mappability = c(0.5,

-  0.1, 0.7), reptiming = NULL, off.target.seqlevels = c("targeted", "all"))

+calculateGCContentByInterval(...)

 }

 \arguments{

-\item{interval.file}{File specifying the intervals. Interval is expected in

-first column in format CHR:START-END.  Instead of a file, a \code{GRanges}

-object can be provided. This allows the use of BED files for example. Note

-that GATK interval files are 1-based (first position of the genome is 1).

-Other formats like BED files are often 0-based. The \code{import} function

-will automatically convert to 1-based \code{GRanges}.}

-

-\item{reference.file}{Reference FASTA file.}

-

-\item{output.file}{Optionally, write GC content file.}

-

-\item{off.target}{Include off-target regions.}

-

-\item{average.target.width}{Split large targets to approximately this size.}

-

-\item{min.off.target.width}{Only include off-target regions of that

-size}

-

-\item{average.off.target.width}{Split off-target regions to that}

-

-\item{off.target.padding}{Pad off-target regions.}

-

-\item{mappability}{Annotate intervals with mappability score. Assumed on a scale

-from 0 to 1, with score being 1/(number alignments). Expected as \code{GRanges}

-object with first meta column being the score. Regions outside these ranges are

-ignored, assuming that \code{mappability} covers the whole accessible genome.}

-

-\item{min.mappability}{\code{double(3)} specifying the minimum mappability score

-for on-target, off-target, and chrY regions in that order. The chrY regions

-are only used for sex determination in \sQuote{PureCN} and are therefore 

-treated differently. Requires \code{mappability}.}

-

-\item{reptiming}{Annotate intervals with replication timing score. Expected as 

-\code{GRanges} object with first meta column being the score.}

-

-\item{off.target.seqlevels}{Controls how to deal with chromosomes/contigs

-found in the \code{reference.file} but not in the \code{interval.file}.}

-}

-\value{

-Returns GC content by interval as \code{GRanges} object.

+\item{...}{Arguments passed to \code{\link{preprocessIntervals}}.}

 }

 \description{

-Uses \code{scanFa} from the Rsamtools package to retrieve GC content of

-intervals in a reference FASTA file. Can optimize intervals for copy

-number calling by tiling long intervals and by including off-target regions.

-This optimization largely follows CNVkit.

-}

-\examples{

-

-reference.file <- system.file("extdata", "ex2_reference.fa", 

-    package="PureCN", mustWork = TRUE)

-interval.file <- system.file("extdata", "ex2_intervals.txt", 

-    package="PureCN", mustWork = TRUE)

-bed.file <- system.file("extdata", "ex2_intervals.bed", 

-    package="PureCN", mustWork = TRUE)

-calculateGCContentByInterval(interval.file, reference.file, 

-    output.file="gc_file.txt")

-

-intervals <- import(bed.file)

-calculateGCContentByInterval(intervals, reference.file, 

-    output.file="gc_file.txt")

-

-}

-\references{

-Talevich et al. (2016). CNVkit: Genome-Wide Copy Number 

-Detection and Visualization from Targeted DNA Sequencing. PLoS Comput Biol.

-}

-\author{

-Markus Riester

+This function was renamed to \code{\link{preprocessIntervals}}.

 }

@@ -26,9 +26,8 @@ callAlterationsFromSegmentation(sampleid, chr, start, end, num.mark = NA,

 \item{gc.gene.file}{A mapping file that assigns GC content and gene symbols

 to each exon in the coverage files. Used for generating gene-level calls.

 First column in format CHR:START-END. Second column GC content (0 to 1).

-Third column gene symbol. This file can be generated with the \sQuote{GATK

-GCContentByInterval} tool or with the

-\code{\link{calculateGCContentByInterval}} function.}

+Third column gene symbol. This file is generated with the 

+\code{\link{preprocessIntervals}} function.}

 \item{fun.focal}{Function for identifying focal amplifications. Defaults to

 \code{\link{findFocal}}.}

@@ -14,9 +14,8 @@ correctCoverageBias(coverage.file, gc.gene.file, output.file = NULL,

 \item{gc.gene.file}{File providing GC content for each exon in the coverage

 files. First column in format CHR:START-END. Second column GC content (0 to

 1).  Third column provides gene symbols, which are optional, but used in

-\code{\link{runAbsoluteCN}} to generate gene level calls. This file can be

-generated with GATK GCContentByInterval tool or with the

-\code{\link{calculateGCContentByInterval}} function.}

+\code{\link{runAbsoluteCN}} to generate gene level calls. This file is 

+generated with the \code{\link{preprocessIntervals}} function.}

 \item{output.file}{Optionally, write file with GC corrected coverage. Can be

 read with the \code{\link{readCoverageFile}} function.}

@@ -46,7 +45,7 @@ coverage <- correctCoverageBias(normal.coverage.file, gc.gene.file)

 }

 \seealso{

-\code{\link{calculateGCContentByInterval}}

+\code{\link{preprocessIntervals}}

 }

 \author{

 Angad Singh, Markus Riester

@@ -227,9 +227,8 @@ are sequencing errors).}

 \item{gc.gene.file}{A mapping file that assigns GC content and gene symbols

 to each exon in the coverage files. Used for generating gene-level calls.

 First column in format CHR:START-END. Second column GC content (0 to 1).

-Third column gene symbol. This file can be generated with the \sQuote{GATK

-GCContentByInterval} tool or with the

-\code{\link{calculateGCContentByInterval}} function.}

+Third column gene symbol. This file is generated with the

+\code{\link{preprocessIntervals}} function.}

 \item{max.dropout}{Measures GC bias as ratio of coverage in AT-rich (GC <

 0.5) versus GC-rich on-target regions (GC >= 0.5). High drop-out might 

similarity index 81%

rename from tests/testthat/test_calculateGCContentByInterval.R

rename to tests/testthat/test_preprocessIntervals.R

@@ -1,4 +1,4 @@

-context("calculateGCContentByInterval")

+context("preprocessIntervals")

 reference.file <- system.file("extdata", "ex2_reference.fa", 

     package = "PureCN", mustWork = TRUE)

@@ -10,7 +10,7 @@ bed.file <- system.file("extdata", "ex2_intervals.bed", package = "PureCN",

 output.file <- tempfile(fileext = ".txt")

 test_that("GC-bias of example reference and intervals (GATK format) matches", {

-    gc <- calculateGCContentByInterval(interval.file, reference.file, 

+    gc <- preprocessIntervals(interval.file, reference.file, 

         output.file = output.file)

     x <- read.delim(output.file, as.is = TRUE)

     expect_equal(x$gc_bias, c(0.4533333, 0.5057143, 0.5733333, 

@@ -23,7 +23,7 @@ test_that("GC-bias of example reference and intervals (BED format) matches", {

     x <- read.delim(output.file, as.is = TRUE)

     intervals <- import(bed.file)

     output.file2 <- tempfile(fileext = ".txt")

-    y <- calculateGCContentByInterval(intervals, reference.file, 

+    y <- preprocessIntervals(intervals, reference.file, 

         output.file = output.file2)

     expect_equal(y$gc_bias, x$gc_bias)

     expect_equal(as.character(y), x$Target)

@@ -38,7 +38,7 @@ test_that("Exceptions happen with wrong input", {

     idata[3, 1] <- "seq2:0-149"

     write.table(idata, file = interval.file2, row.names = FALSE, 

         quote = FALSE)

-    expect_error(calculateGCContentByInterval(interval.file2, 

+    expect_error(preprocessIntervals(interval.file2, 

         reference.file),

         "Interval coordinates should start at 1, not at 0")

     file.remove(interval.file2)

@@ -50,17 +50,17 @@ test_that("reptiming annotated correctly", {

     reptiming <- import(reptiming.file)

     intervals <- import(bed.file)

-    gr <- calculateGCContentByInterval(intervals, reference.file, 

+    gr <- preprocessIntervals(intervals, reference.file, 

         reptiming = reptiming)

     expect_equal(c(17.5, 11.0, 50, 10.0, 10.0), gr$reptiming)

 })

 test_that("Offtarget settings work as expected", {

-    gc <- calculateGCContentByInterval(interval.file, reference.file, 

+    gc <- preprocessIntervals(interval.file, reference.file, 

         off.target = TRUE, min.off.target.width = 2, off.target.padding = -2)

     expect_equal(length(gc), 11)

     intervals <- import(bed.file)

-    gc2 <- calculateGCContentByInterval(gc, reference.file)

+    gc2 <- preprocessIntervals(gc, reference.file)

     expect_equal(start(gc2), start(gc))

     expect_equal(end(gc2), end(gc))

     expect_equal(gc2$mappability, gc$mappability)

@@ -69,14 +69,14 @@ test_that("Offtarget settings work as expected", {

         mappability.file <- system.file("extdata", "ex2_mappability.bigWig", 

             package = "PureCN", mustWork = TRUE)

         mappability <- import(mappability.file)

-        gcMap <- calculateGCContentByInterval(intervals, reference.file, 

+        gcMap <- preprocessIntervals(intervals, reference.file, 

             mappability = mappability)

         expect_equal(gcMap$mappability, c(1, 1, 0.7, 1, 1), tolerance = 0.001)

     }

     mappability.file <- system.file("extdata", "ex2_mappability.bed", 

         package = "PureCN", mustWork = TRUE)

     mappability <- import(mappability.file)

-    gcMap <- calculateGCContentByInterval(intervals, reference.file, 

+    gcMap <- preprocessIntervals(intervals, reference.file, 

         mappability = mappability)

     expect_equal(gcMap$mappability, c(1, 1, 0.7, 1, 1), tolerance = 0.001)

     reference.file <- system.file("extdata", "ex3_reference.fa", 

@@ -84,20 +84,20 @@ test_that("Offtarget settings work as expected", {

     bed.file3 <- system.file("extdata", "ex3_intervals.bed", 

         package = "PureCN", mustWork = TRUE)

     intervals3 <- import(bed.file3)

-    x <- calculateGCContentByInterval(intervals3, reference.file)

+    x <- preprocessIntervals(intervals3, reference.file)

     expect_equal(x$gc_bias, c(0.4533333, 0.5057143, 0.5733333,

         0.48, 0.36), tolerance = 0.001)

     seqlevelsStyle(intervals3) <- "NCBI"

-    x <- calculateGCContentByInterval(intervals3, reference.file)

+    x <- preprocessIntervals(intervals3, reference.file)

     expect_equal(x$gc_bias, c(0.4533333, 0.5057143, 0.5733333, 

         0.48, 0.36), tolerance = 0.001)

-    expect_error(calculateGCContentByInterval(intervals, reference.file),

+    expect_error(preprocessIntervals(intervals, reference.file),

         "Chromosome naming style of interval file")

     mappability.file3 <- system.file("extdata", "ex3_mappability.bed", 

         package = "PureCN", mustWork = TRUE)

     mappability3 <- import(mappability.file3)

     seqlevelsStyle(mappability3) <- "NCBI"

-    x <- calculateGCContentByInterval(intervals3, reference.file, 

+    x <- preprocessIntervals(intervals3, reference.file, 

         mappability = mappability3)

     expect_equal(x$gc_bias, c(0.4533333, 0.5057143, 0.5733333, 

         0.48, 0.36), tolerance = 0.001)

@@ -132,7 +132,7 @@ reads can be used by the segmentation function.

 It further annotates targets by GC-content (how coverage is normalized is

 described later in Section~\ref{secgcbias}). 

-\Biocpkg{PureCN} provides the \Rfunction{calculateGCContentByInterval} 

+\Biocpkg{PureCN} provides the \Rfunction{preprocessIntervals} 

 function:

 <<examplegc>>=

@@ -146,7 +146,7 @@ mappability.file <- system.file("extdata", "ex2_mappability.bigWig",

 intervals <- import(bed.file)

 mappability <- import(mappability.file)

-calculateGCContentByInterval(intervals, reference.file, 

+preprocessIntervals(intervals, reference.file, 

     mappability=mappability, output.file = "ex2_gc_file.txt")

 @

@@ -159,7 +159,7 @@ symbols using the \Rfunction{annotateTargets} function.

 The \Rfunction{calculateBamCoverageByInterval} function can be used to generate

 the required coverage data from BAM files. All we need to do is providing the

-desired intervals (as generated by \Rfunction{calculateGCContentByInterval}):

+desired intervals (as generated by \Rfunction{preprocessIntervals}):

 <<examplecoverage>>=

 bam.file <- system.file("extdata", "ex1.bam", package="PureCN",