Merge branch 'master' into devel

* master: fix typo document issue 160 in readMSData2 acq num warning add a warning if acq num not sorted - close issue #160 fix confusing typo in news update readme injection time is now added by default in readMSData2 mention v2.0 and readMSData2 in main vignette suggest on latest msdata fix bug in addInjectionTime rename to addInjectionTime new injectionTime method (see issue #159) - code in previous commit specify pattern when getting msdata::proteomics files new gh devel version From: Laurent <[email protected]> git-svn-id: https://hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/MSnbase@122003 bc3139a8-67e5-0310-9ffc-ced21a209358
lgatto · Oct 6, 2016 · e108d6b · e108d6b
1 parent 1ffc6b8
commit e108d6b
Show file tree

Hide file tree

Showing 13 changed files with 50 additions and 12 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: MSnbase
 Title: Base Functions and Classes for MS-based Proteomics
-Version: 1.99.2
+Version: 1.99.3
 Description: Basic plotting, data manipulation and processing
 	     of MS-based Proteomics data.
 Authors@R: c(person(given = "Laurent", family = "Gatto",
@@ -63,6 +63,7 @@ Imports:
     lattice,
     ggplot2,
     S4Vectors,
+    XML,
     Rcpp
 Suggests:
     testthat,
@@ -75,7 +76,7 @@ Suggests:
     Rdisop,
     pRoloc,
     pRolocdata (>= 1.7.1),
-    msdata,
+    msdata (>= 0.12.2),
     roxygen2,
     rgl,
     rpx,
@@ -85,7 +86,6 @@ Suggests:
     imputeLCMD,
     norm,
     gplots,
-    XML,
     shiny
 LinkingTo: Rcpp
 License: Artistic-2.0

diff --git a/NAMESPACE b/NAMESPACE
@@ -12,6 +12,9 @@ import(impute)
 import(lattice)
 import(methods)
 
+importFrom(XML, xmlInternalTreeParse, getDefaultNamespace, xpathApply,
+           xmlAttrs)
+
 importFrom(graphics, abline, axis, grid, legend, lines, points, text,
            barplot, layout, par)
 importFrom(stats, ave, median, medpolish, quantile, reorder, setNames,
@@ -156,7 +159,6 @@ exportMethods(updateObject,
               "sampleNames<-",
               fileNames,
               msInfo,
-              isolationWindow,
               expemail, exptitle,
               ionSource, ionSourceDetails,
               analyser, analyzer,

diff --git a/NEWS b/NEWS
@@ -1,3 +1,8 @@
+CHANGES IN VERSION 1.99.3
+-------------------------
+ o Injection time is now added to the header when reading mzML files
+   using readMSData2 (see issue #159) <2016-10-04 Tue>
+
 CHANGES IN VERSION 1.99.2
 -------------------------
  o Added isolationWindow,MSnExp method <2016-09-23 Fri>

diff --git a/R/AllGenerics.R b/R/AllGenerics.R
@@ -140,3 +140,5 @@ setGeneric("filterRt", function (object, ...) standardGeneric("filterRt"))
 setGeneric("filterMsLevel", function (object, ...) standardGeneric("filterMsLevel"))
 setGeneric("filterFile", function (object, ...) standardGeneric("filterFile"))
 setGeneric("filterAcquisitionNum", function (object, ...) standardGeneric("filterAcquisitionNum"))
+
+## isolationWindow generic is in mzR
diff --git a/R/header.R b/R/header.R
@@ -0,0 +1,13 @@
+## These data should eventually be extracted in mzR and added to the
+## default header
+
+injectionTimeFromFile1 <- function(f) {
+    xvals <- vector("list", length = length(f))
+    doc <- XML::xmlInternalTreeParse(f)
+    namespaceDef <- XML::getDefaultNamespace(doc)
+    ns <- c(x = namespaceDef[[1]]$uri)
+    x <- XML::xpathApply(doc, "//x:cvParam[@accession = 'MS:1000927']",
+                         namespaces = ns, xmlAttrs)
+    xvals <- sapply(x, "[", "value")
+    as.numeric(unlist(xvals))
+}
diff --git a/R/readMSData2.R b/R/readMSData2.R
@@ -46,10 +46,18 @@ readMSData2 <- function(files,
                         centroided = rep(as.logical(NA), nrow(fdData)),
                         smoothed = rep(as.logical(smoothed.), nrow(fdData)),
                         fdData, stringsAsFactors = FALSE)
+        injt <- injectionTimeFromFile1(f)
+        if (is.numeric(injt) && length(injt) == nrow(fdData))
+            fdData$injectionTime <- injt
         ## Order the fdData by acquisitionNum to force use of acquisitionNum
         ## as unique ID for the spectrum (issue #103). That way we can use
         ## the spIdx (is the index of the spectrum within the file) for
         ## subsetting and extracting.
+        if (!all(sort(fdData$acquisitionNum) == fdData$acquisitionNum))
+            warning(paste("Unexpected acquisition number order detected.",
+                          "Please contact the maintainers or open an issue",
+                          "on https://github.com/lgatto/MSnbase.",
+                          sep = "\n")) ## see issue #160
         fdData <- fdData[order(fdData$acquisitionNum), ]
         featureDataList <- c(featureDataList, list(fdData))
         mzR::close(msdata)

diff --git a/man/OnDiskMSnExp-class.Rd b/man/OnDiskMSnExp-class.Rd
@@ -580,7 +580,7 @@ mzsAfter <- intensity(odmse)
 head(lengths(mzs))
 head(lengths(mzsAfter))
 
-f <- msdata::proteomics(full.names = TRUE)
+f <- msdata::proteomics(full.names = TRUE, pattern = "TMT_Erwinia")
 odmse <- readMSData2(f)
 validObject(odmse)
 odmse[[1]]

diff --git a/tests/testthat/test_MSnExp.R b/tests/testthat/test_MSnExp.R
@@ -326,7 +326,7 @@ test_that("Noise estimation MSnExp", {
 })
 
 test_that("isolation window", {
-    f <- msdata::proteomics(full.names = TRUE)
+    f <- msdata::proteomics(full.names = TRUE, pattern = "TMT_Erwinia")
     i1 <- isolationWindow(f, unique = FALSE)
     i2 <- isolationWindow(readMSData2(f), unique = FALSE)
     i3 <- isolationWindow(readMSData(f), unique = FALSE)

diff --git a/tests/testthat/test_OnDiskMSnExp.R b/tests/testthat/test_OnDiskMSnExp.R
@@ -45,7 +45,7 @@ test_that("OnDiskMSnExp constructor", {
 })
 
 test_that("Coercion to MSnExp", {
-    f <- msdata:::proteomics(full.names = TRUE)
+    f <- msdata:::proteomics(full.names = TRUE, pattern = "TMT_Erwinia")
     x <- readMSData2(f, verbose = FALSE)
     y <- readMSData(f, msLevel. = 2, centroided. = NA, verbose = FALSE)
     expect_error(as(x, "MSnExp"))

diff --git a/tests/testthat/test_centroided.R b/tests/testthat/test_centroided.R
@@ -1,5 +1,5 @@
 test_that("centroided accessor with/without na.fail", {
-    f <- msdata::proteomics(full.names = TRUE)
+    f <- msdata::proteomics(full.names = TRUE, pattern = "TMT_Erwinia")
     x <- readMSData(f)
     x2 <- readMSData2(f)
     ## in-mem, single spectrum

diff --git a/tests/testthat/test_readMSData2.R b/tests/testthat/test_readMSData2.R
@@ -1,7 +1,7 @@
 context("readMSData2")
 
 test_that("msLevel set correctly", {
-    f <- msdata::proteomics(full.names = TRUE)
+    f <- msdata::proteomics(full.names = TRUE, pattern = "TMT_Erwinia")
     x <- readMSData2(f, verbose = FALSE)
     expect_equivalent(centroided(x), rep(NA, length(x)))
     x <- readMSData2(f, centroided = TRUE, verbose = FALSE)

diff --git a/vignettes/MSnbase-demo.Rnw b/vignettes/MSnbase-demo.Rnw
@@ -109,7 +109,7 @@ format\footnote{%%
 \paragraph{Raw data} The \texttt{XML}-based formats, \texttt{mzXML}
 \cite{Pedrioli2004}, \texttt{mzData} \cite{Orchard2007} and
 \texttt{mzML} \cite{Martens2010} can be imported with the
-\Rfunction{readMSData} function, as illstrated below (see
+\Rfunction{readMSData} function, as illustrated below (see
 \Rfunction{?readMSData} for more details).
 
 
@@ -120,13 +120,21 @@ rawdata <- readMSData(file, msLevel = 2, verbose = FALSE)
 @
 %% $
 
-Either MS1 or MS2 spectra can be loaded at a time by setting the
+Only spectra of a give MS level can be loaded at a time by setting the
 \texttt{msLevel} parameter accordingly.  In this document, we will use
 the \Robject{itraqdata} data set, provided with \Biocpkg{MSnbase}. It
 includes feature metadata, accessible with the \Rfunction{fData}
 accessor.  The metadata includes identification data for the
 \Sexpr{length(itraqdata)} MS2 spectra.
 
+\paragraph{MSnbase 2.0 } The new major version of \Biocpkg{MSnbase}
+uses a new \textit{on-disk} data storage model (see the
+\textit{benchmarking} vignette for more details). The new data backend
+is compatible with the orignal \textit{in-memory} model. To make use
+of the new infrastructure, read your raw data using
+\Rfunction{readMSData2}, rather than \Rfunction{readMSData}. All
+existing operations work irrespective of the backend.
+
 \paragraph{Peak lists} Peak lists can often be exported after spectrum
 processing from vendor-specific software and are also used as input to
 search engines.  Peak lists in \texttt{mgf} format can be imported

diff --git a/vignettes/benchmarking.Rmd b/vignettes/benchmarking.Rmd
@@ -38,7 +38,7 @@ distributed with the `r Biocexptpkg("msdata")` package
 
 ```{r msdata}
 library("msdata")
-f <- msdata::proteomics(full.names = TRUE)
+f <- msdata::proteomics(full.names = TRUE, pattern = "TMT_Erwinia")
 basename(f)
 ```