Fast statistical alignment - PubMed (original) (raw)

Fast statistical alignment

Robert K Bradley et al. PLoS Comput Biol. 2009 May.

Abstract

We describe a new program for the alignment of multiple biological sequences that is both statistically motivated and fast enough for problem sizes that arise in practice. Our Fast Statistical Alignment program is based on pair hidden Markov models which approximate an insertion/deletion process on a tree and uses a sequence annealing algorithm to combine the posterior probabilities estimated from these models into a multiple alignment. FSA uses its explicit statistical model to produce multiple alignments which are accompanied by estimates of the alignment accuracy and uncertainty for every column and character of the alignment--previously available only with alignment programs which use computationally-expensive Markov Chain Monte Carlo approaches--yet can align thousands of long sequences. Moreover, FSA utilizes an unsupervised query-specific learning procedure for parameter estimation which leads to improved accuracy on benchmark reference alignments in comparison to existing programs. The centroid alignment approach taken by FSA, in combination with its learning procedure, drastically reduces the amount of false-positive alignment on biological data in comparison to that given by other methods. The FSA program and a companion visualization tool for exploring uncertainty in alignments can be used via a web interface at http://orangutan.math.berkeley.edu/fsa/, and the source code is available at http://fsa.sourceforge.net/.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1

Figure 1. Overview of the components constituting the FSA alignment program.

The algorithms that are used in each component are highlighted in the accompanying boxes. The bold arrows show the simplest mode of use for FSA, where posterior probabilities are calculated directly using default parameters for all pairs of sequences and the optional steps of anchor finding and iterative refinement are omitted.

Figure 2

Figure 2. The default Pair HMM used by FSA.

By default FSA uses a Pair HMM with two sets of Insert (I) and Delete (D) states to generate a two-component geometric mixture distribution. FSA can optionally use a three-state HMM, which has only one set of Insert and Delete states. M is a Match state emitting aligned characters.

Figure 3

Figure 3. Two alignments (left and right) which make the same homology statements and therefore are both represented by the same POSET (center).

“The mathematics of distance-based alignment” in Text S1 discusses this view of alignments as POSETs. The alignment on the right minimizes the number of gap-open events and as such is appropriate for analyses such as inferring parsimonious indel frequencies across a clade. Alignments are displayed with TeXshade .

Figure 4

Figure 4. Schematic overview of FSA's parallelization strategy on a computer cluster.

For large input sizes, a disk-based database may be used to store some of the primary data structures and reduce memory usage.

Figure 5

Figure 5. The Java GUI allows users to visualize the estimated alignment accuracy under FSA's statistical model.

FSA's alignment is colored according the expected accuracy under FSA's statistical model (top) as well as according to the “true” accuracy (bottom) given from a comparison between FSA's alignment and the reference structural alignment. It is clear from inspection that accuracies estimated under FSA's statistical model correspond closely to the true accuracies. Sequences are from alignment BBS12030 in the RV12 dataset of BAliBASE 3 .

Similar articles

Cited by

References

    1. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Research. 1994;22:4673–4680. - PMC - PubMed
    1. Larkin M, Blackshields G, Brown N, Chenna R, McGettigan P, et al. Clustal Wand Clustal X version 2.0. Bioinformatics. 2007;23:2947–2948. - PubMed
    1. Edgar R, Batzoglou S. Multiple sequence alignment. Curr Opin Struct Biol. 2006;16:368–373. - PubMed
    1. Lunter G, Rocco A, Mimouni N, Heger A, Caldeira A, et al. Uncertainty in homology inferences: assessing and improving genomic sequence alignment. Genome Research 2007 - PMC - PubMed
    1. Wong K, Suchard M, Huelsenbeck J. Alignment uncertainty and genomic analysis. Science. 2008;319:473–476. - PubMed

Publication types

MeSH terms

LinkOut - more resources