BIOINFORMATICS APPLICATIONS NOTE Vol. 19 no. 12 2003, pages 1592–1593DOI: 10.1093/bioinformatics/btg197
QAlign: quality-based multiple alignments with
dynamic phylogenetic analysis
M. Sammeth 1, 2, 3,∗,†, J. Rothga¨nger 4, W. Esser 1, J. Albert 1,
J. Stoye 3 and D. Harmsen 5
1Department of Computer Science II, University of Wu¨rzburg, 97074 Wu¨rzburg,
Germany, 2International Graduate School of Bioinformatics, Bielefeld University,
33594 Bielefeld, Germany, 3Genome Informatics, Department of Technology,
Bielefeld University, 33594 Bielefeld, Germany, 4RIDOM bioinformatics, 97082
Wu¨rzburg, Germany and 5Institute for Hygiene, University of Mu¨nster, 48149
Mu¨nster, Germany
Received on December 11, 2002; revised on February 12, 2003; accepted on March 4, 2003
ABSTRACT
Summary: Integrating different alignment strategies, a
layout editor and tools deriving phylogenetic trees in
a ‘multiple alignment environment’ helps to investigate
and enhance results of multiple sequence alignment by
hand. QAlign combines algorithms for fast progressive and
accurate simultaneous multiple alignment with a versatile
editor and a dynamic phylogenetic analysis in a convenient
graphical user interface.
Availability: QAlign is freely available over the internet
at http://www.ridom.de/qalign/. The platform-independent
JAVA technology used provides distributions for various
operating systems and hardware architectures.
Contact: qalign@ridom.de
INTRODUCTION
The correct alignment of multiple DNA and protein
sequences is a fundamental problem in computational
biology. Results produced by the commonly used progres-
sive multiple alignment methods can be obtained rapidly
but they are highly dependent on the degree of similarity
of the input. Simultaneous alignment algorithms syn-
chronize the information in all sequences to construct
the multiple alignment and are therefore more sensitive.
However, even these optimal alignment layouts may need
some manual editing. Furthermore, downstream analyses
(e.g. methods to derive phylogenetic trees) are linked
dynamically to the multiple alignment. Thus, a stronger
interaction between the creation of the alignment and the
phylogenetic analysis enables evolutionary trees of high
quality to be found.
∗To whom correspondence should be addressed.
† Present address: Genome Informatics, Department of Technology,
Bielefeld University, 33594 Bielefeld, Germany.
IMPLEMENTATION
Due to its modular and layered structure, our program
QAlign may easily be extended to support additional
algorithms for both, multiple alignment and phylogenetic
reconstruction. Herein we outline the features included in
the current version.
Multiple alignment algorithms
QAlign is a new graphical environment integrating multi-
ple features in the construction of the best multiple align-
ment for a specific set of sequences (FASTA and MSF
sequence format supported). The algorithm monitor con-
trols the construction of the multiple alignment where a
fast progressive or a more accurate simultaneous approach
may be chosen to align the sequences or parts of them
(see Fig. 1 right). The heuristics used in the progressive
approach (QAlign uses the variant of the MSA protocol,
Gupta et al., 1995) of global multiple sequence alignment
allow the alignment of even very large data sets. However,
the drawback is that the resulting alignment is a fast ap-
proximation of the solution (McClure et al., 1994; Hick-
son et al., 2000).
In addition, QAlign contains an efficient and stable re-
implementation of the NCBI’s MSA (multiple sequence
alignment) program (Gupta et al., 1995). This is based on
the simultaneous alignment strategy, an exact algorithm
capable of finding the optimal mathematical solution. In
addition to the optimizations used in MSA, the divide-
and-conquer algorithm DCA (To¨nges et al., 1996; Stoye et
al., 1997) was used to achieve the simultaneous alignment
of larger data sets. The desired quality-time tradeoff ratio
for simultaneous alignment construction can be balanced
by a slider. Both the progressive and the simultaneous
alignment strategy may be used in a complementary
manner on the same alignment layout.
1592 Bioinformatics 19(12) c© Oxford University Press 2003; all rights reserved.
by guest on Septem
ber 17, 2011
bioinform
atics.oxfordjournals.org
D
ow
nloaded from
QAlign
Fig. 1. The graphical user interface of QAlign: the neightbour joining tree is updated dynamically (top) different algorithms and their
parameters are accessible by the algorithm monitor (centre-right) and context menus support the editing functions for each block (bottom).
Alignment editor features
After aligning the sequences, the graphical editor of
QAlign provides features to analyse the result and modify
the multiple alignment layout (as in Fig. 1, bottom).
Gaps may be inserted or deleted and marked blocks
may be moved within the alignment providing that the
aligned sequences have the same length. An immediately
updated consensus sequence with coloured bars shows
the matching ratio of each column. These bars represent
the conservation of different clusters across the alignment
layout. They are also displayed as a bird’s eye view under-
neath a scrollbar thus allowing easy navigation to areas of
low similarity. A secondary view is provided which may
be used to extend the editor capabilities on one alignment
or to compare two different multiple alignment layouts.
Dynamic phylogenetic analysis
A dynamic phylogenetic tree view makes visible the con-
sequences of a change in the alignment with regard to the
phylogenetic relationship (see Fig. 1, top) where branch
lengths may change and nodes may swap according to
the neighbour-joining method (Saitou and Nei, 1987). The
tree may also be bootstrapped at any time to reveal its cur-
rent stability. Thus, a phylogenetic reflection of the dy-
namics of the multiple alignment layout is obtained.
A variety of visual rearrangements is provided for
the tree (e.g. subtrees may be collapsed or rearranged).
Finally, the phylogenetic tree may be exported, either to
a vectorial data format for drawing tools (SVG) or to a
common format used by tree plotters (Newick).
CONCLUSION
QAlign provides a practical solution for the creation of
refined multiple alignments: layouts produced by various
algorithms may be used as a starting-point for changes
done by hand, while the phylogenetic consequences are
visualised on the fly. Furthermore, the comparison of
multiple alignments is made easier because of the two
alignment views integrated in the user interface of QAlign.
REFERENCES
Gupta,S.K., Kececioglu,J.D. and Scha¨ffer,A.A. (1995) Improving
the practical space and time efficiency of the shortest-paths
approach to sum-of-pairs multiple sequence alignment. J. Comp.
Biol., 2, 459–472.
Hickson,R.E., Simon,C. and Perrey,S.W. (2000) The performance
of several multiple-sequence alignment programs in relation to
secondary-structure features for an rRNA sequence. Mol. Biol.
Evol., 17, 530–539.
McClure,M.A., Vasi,T.K. and Fitch,W.M. (1994) Comparative anal-
ysis of multiple protein-sequence alignment methods. Mol. Biol.
Evol., 11, 571–592.
Saitou,N. and Nei,M. (1987) The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Mol. Biol. Evol.,
4, 406–425.
Stoye,J., Moulton,V. and Dress,A.W. (1997) DCA: an efficient im-
plementation of the divide-and-conquer approach to simultane-
ous multiple sequence alignment. Comput. Appl. Biosci., 13,
625–626.
To¨nges,U., Perrey,S.W., Stoye,J. and Dress,A.W.M. (1996) A
general method for fast multiple sequence alignment. Gene, 172,
GC33–GC41.
1593
by guest on Septem
ber 17, 2011
bioinform
atics.oxfordjournals.org
D
ow
nloaded from
