@@ -1,8 +1,8 @@

 Package: msa

 Type: Package

 Title: Multiple Sequence Alignment

-Version: 1.3.3

-Date: 2016-02-25

+Version: 1.3.4

+Date: 2016-04-06

 Author: Enrico Bonatesta, Christoph Horejs-Kainrath, Ulrich Bodenhofer

 Maintainer: Ulrich Bodenhofer <bodenhofer@bioinf.jku.at>

 Description: This package provides a unified R/Bioconductor interface to the

@@ -27,7 +27,7 @@ LazyLoad: yes

 Collate: AllClasses.R AllGenerics.R params-methods.R version-methods.R

         helperFunctions.R inputChecks.R convertRows.R msaPrettyPrint.R

         print-methods.R show-methods.R msa.R msaMuscle.R msaClustalW.R

-        msaClustalOmega.R msaConvert.R

+        msaClustalOmega.R msaConvert.R msaCheckNames.R

 biocViews: MultipleSequenceAlignment, Alignment, MultipleComparison,

         Sequencing

 NeedsCompilation: yes

@@ -7,9 +7,10 @@ import(Biostrings)

 import(S4Vectors)

 import(IRanges)

 importFrom(tools, texi2dvi)

+importFrom(utils, capture.output)

 export(msa, msaMuscle, msaClustalW, msaClustalOmega, msaPrettyPrint,

-       msaConvert)

+       msaConvert, msaCheckNames)

 exportClasses(MsaDNAMultipleAlignment, MsaRNAMultipleAlignment,

               MsaAAMultipleAlignment, MsaMetaData)

new file mode 100644

@@ -0,0 +1,20 @@

+msaCheckNames <- function(x, replacement=" ", verbose=TRUE)

+{

+    if (!is(x, "MultipleAlignment"))

+        stop("x must be a multiple alignment object")

+

+    out <- x

+

+    pattern <- "[^a-zA-Z0-9,;:.?!\\-\\(\\)\\'\" ]"

+

+    if (length(grep(pattern, rownames(x), perl=TRUE)) > 0)

+    {

+        if (verbose)

+            message("sequence names contain invalid characters")

+

+        rownames(out) <- gsub(pattern, replacement, rownames(x),

+                              perl=TRUE)

+    }

+

+    invisible(out)

+}

@@ -1,6 +1,13 @@

 Change history of package msa:

 ==============================

+Version 1.3.4:

+- added function for checking and fixing sequence names for

+  possibly problematic characters that could lead to LaTeX

+  errors when using msaPrettyPrint()

+- corresponding changes in documentation

+- minor namespace fix

+

 Version 1.3.3:

 - added function for converting multiple sequence alignments for

   use with other sequence alignment packages

new file mode 100644

@@ -0,0 +1,91 @@

+\name{msaCheckNames}

+\alias{msaCheckNames}

+\title{Check and fix sequence names}

+\description{

+  This function checks and fixed sequence names of multiple

+  alignment objects if they contain characters

+  that might lead to LaTeX problems when using

+  \code{\link{msaPrettyPrint}}.

+}

+\usage{

+    msaCheckNames(x, replacement=" ", verbose=TRUE)

+}

+\arguments{

+  \item{x}{an object of class \code{\linkS4class{MultipleAlignment}}

+    (which includes objects of classes

+    \code{\linkS4class{MsaAAMultipleAlignment}},

+    \code{\linkS4class{MsaDNAMultipleAlignment}}, and

+    \code{\linkS4class{MsaRNAMultipleAlignment}})}

+  \item{replacement}{a character string specifying with which

+    character(s) potentially problematic characters should be replaced.}

+  \item{verbose}{if \code{TRUE} (default), a warning message is shown

+    if potentially problematic characters are found. Otherwise,

+    the function silently replaces these characters (see details below).}

+}

+\details{

+  The \pkg{Biostrings} package does not impose any restrictions on

+  the names of sequences. Consequently, \pkg{msa} also allows all

+  possible ASCII strings as sequence (row) names in multiple alignments.

+  As soon as \code{\link{msaPrettyPrint}} is used for pretty-printing

+  multiple sequence alignments, however, the sequence names are

+  interpreted as plain LaTeX source code. Consequently, LaTeX errors

+  may arise because of characters or words in the sequence names that LaTeX

+  does not or cannot interpret as plain text correctly. This

+  particularly includes appearances of special characters and backslash

+  characters in the sequence names.

+

+  The \code{msaCheckNames} function takes a multiple alignment object

+  and checks sequence names for possibly problematic characters, which

+  are all characters but letters (upper and lower case), digits,

+  spaces, commas, colons, semicolons, periods, question and exclamation

+  marks, dashes, braces, single quotes, and double quotes.

+  All other characters are

+  considered problematic. The function allows for both checking and

+  fixing the sequence names. If called with \code{verbose=TRUE}

+  (default), the function prints a warning if a problematic character is

+  found. At the same time, regardless of the \code{verbose} argument,

+  the function invisibly returns a copy of \code{x} in whose sequence

+  names all problematic characters have been replaced by the string

+  that is supplied via the \code{replacement} argument (the default is

+  a single space).

+

+  In any case, the best solution is to check sequence names carefully

+  and to avoid problematic sequence names from the beginning.

+}

+\value{

+  The function invisibly returns a copy of the argument \code{x}

+  (therefore, an object of the same class as \code{x}), but

+  with modified sequence/row names (see details above).

+}

+\author{Ulrich Bodenhofer <msa@bioinf.jku.at>

+}

+\references{

+  \url{http://www.bioinf.jku.at/software/msa}

+  

+  U. Bodenhofer, E. Bonatesta, C. Horejs-Kainrath, and S. Hochreiter

+  (2015). msa: an R package for multiple sequence alignment. 

+  \emph{Bioinformatics} \bold{31}(24):3997-3999. DOI:

+  \href{http://dx.doi.org/10.1093/bioinformatics/btv494}{10.1093/bioinformatics/btv494}.

+}

+\seealso{\code{\link{msaPrettyPrint}},

+  \code{\linkS4class{MsaAAMultipleAlignment}},

+  \code{\linkS4class{MsaDNAMultipleAlignment}},

+  \code{\linkS4class{MsaRNAMultipleAlignment}}

+}

+\examples{

+## create toy example

+mySeqs <- DNAStringSet(c("ACGATCGATC", "ACGACGATC", "ACGATCCCCC"))

+names(mySeqs) <- c("Seq. #1", "Seq. \\2", "Seq. ~3")

+

+## perform multiple alignment

+myAlignment <- msa(mySeqs)

+myAlignment

+

+## check names

+msaCheckNames(myAlignment)

+

+## fix names

+myAlignment <- msaCheckNames(myAlignment, replacement="", verbose=FALSE)

+myAlignment

+}

+\keyword{manip}

@@ -666,10 +666,28 @@ the \shade\ package must be loaded in the preamble:

 \end{verbatim}

 \end{quote}

+\subsection{Sequence Names}

+

+The \verb+Biostrings+ package does not impose any restrictions on the names of

+sequences. Consequently, \MSA\ also allows all possible ASCII strings as

+sequence (row) names in multiple alignments. As soon as \verb+msaPrettyPrint()+

+is used for pretty-printing multiple sequence alignments, however, the sequence

+names are interpreted as plain \LaTeX\ source code. Consequently, \LaTeX\ errors

+may arise because of characters or words in the sequence names that \LaTeX\

+does not or cannot interpret as plain text correctly. This particularly includes

+appearances of special characters and backslash characters in the sequence names.

+

+The \MSA\ package offers a function \verb+msaCheckNames()+ which allows for finding

+and replacing potentially problematic characters in the sequence names of

+multiple alignment objects (see \verb+?msaCheckNames+). However, the best solution

+is to check sequence names carefully and to avoid problematic sequence names from the beginning.

+Note, moreover, that too long sequence names will lead to less appealing outputs,

+so users are generally advised to consider sequence names carefully.

+

 \subsection{Further Caveats}

 \begin{itemize}

-  \item Note that \verb+texi2dvi()+ and \verb+ttexi2pdf()+ always

+  \item Note that \verb+texi2dvi()+ and \verb+texi2pdf()+ always

     save the resulting DVI/PDF files to the current working directory,

     even if the \LaTeX\ source file is in a different directory.

     That is also the reason why the temporary file is created in the

@@ -764,6 +782,13 @@ bibliography below).

 \section{Change Log}

 \begin{description}

+\item[Version 1.3.4:] \mbox{ }  \begin{itemize}

+   \item added function for checking and fixing sequence names for

+     possibly problematic characters that could lead to \LaTeX\ errors

+     when using \verb+msaPrettyPrint()+

+   \item corresponding changes in documentation

+   \item minor namespace fix

+  \end{itemize}

 \item[Version 1.3.3:] \mbox{ }  \begin{itemize}

    \item added function for converting multiple sequence alignments for

      use with other sequence alignment packages