From: Jim Procter <jprocter@issues.jalview.org>
Date: Tue, 29 Nov 2011 07:17:01 +0000 (+0000)
Subject: completed table of jabaws alignment services useful for DNA
X-Git-Tag: Version_1.9.2~71
X-Git-Url: http://source.jalview.org/gitweb/?a=commitdiff_plain;h=a580343475d91b927a1ed6a48d7d8c47baaeb96a;p=jalview-manual.git

completed table of jabaws alignment services useful for DNA
---

diff --git a/TheJalviewTutorial.tex b/TheJalviewTutorial.tex
index 5a5986f..339ed2e 100644
--- a/TheJalviewTutorial.tex
+++ b/TheJalviewTutorial.tex
@@ -2662,13 +2662,14 @@ score when aligning two nucleotide sequences.
 \subsubsection{Aligning Nucleic Acid Sequences}
 
 Jalview only has limited knowledge of the capabilities of the programs that
-are made available to it {\sl via} web services. The table below shows which
-alignment programs are most appropriate for nucleotide alignment. 
+are made available to it {\sl via} web services, so it is up to you, the user,
+to decide which service to use when working with nucleic acid sequences. The
+table below shows which alignment programs are appropriate for nucleotide alignment.
 
 \begin{table}{}
 \centering
 \begin{tabular}{|l|c|l|}
-Program& RNA support& Notes\\
+Program& NA support& Notes\\
 \hline
 ClustalW& Yes&\begin{minipage}[f]{3in}
 Default is to autodetect nucleotide
@@ -2676,18 +2677,41 @@ sequences. Editable parameters include nucleotide substitution matrices and
 distance metrics.
 \end{minipage}
 \\
+\hline
+Muscle& Yes (treat U as T)&\begin{minipage}[f]{3in}
+Default is to autodetect nucleotide
+sequences. Editable parameters include nucleotide substitution matrices and
+distance metrics.
+\end{minipage}
+\\
+\hline
+MAFFT& Yes&\begin{minipage}[f]{3in}
+Will autodetect nucleotide sequences and use a hardwired substitution model
+(all amino-acid sequence related parameters are ignored). Unknown whether
+substitution model treats Uracil specially.
+\end{minipage}
+\\
+\hline
+ProbCons& No&\begin{minipage}[f]{3in}
+ProbCons has no special support for aligning nucleotide sequences. Whilst an
+alignment will be returned, it is unlikely to be reliable.
+\end{minipage}
+\\
+\hline
+T-COFFEE& Yes&\begin{minipage}[f]{3in}
+Sequence type is automatically detected and an appropriate
+parameter set used as required. A range of nucleotide specific
+score models are available.\end{minipage}
+\\
 \end{tabular}
-\caption{JABAWS Alignment programs suitable for aligning nucleic acid
-sequences.}
+\caption{{\bf JABAWS Alignment programs suitable for aligning nucleic acid
+sequences.} All JABAWS alignment services will return an alignment if provided
+with RNA or DNA sequences, with varying reliability. No programs include
+support for taking RNA secondary structure prediction into account when
+aligning sequences.}
 \label{nucleomsatools}
 \end{table}
 
-In particular,
-only the ClustalW and MAFFT programs will successfuly recognise and align nucleic acid sequences. MAFFT will also choose an appropriate parameter model. Whilst Muscle may appear to align DNA, it simply treats the base symbols as
-amino-acids, often leading to a poor quality alignment. Furthermore, it will
-almost certainly fail to align RNA containing Uracil bases, since `U' is not a
-valid one-letter amino acid code.
-
 \subsection{Translate cDNA}
 
 The {\sl Calculations $\Rightarrow$ Translate cDNA} function in the alignment