Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013;5(8):1470-84.
doi: 10.1093/gbe/evt105.

Alignment-free genome tree inference by learning group-specific distance metrics

Kaustubh R Patil  1 , Alice C McHardy
Affiliations

Alignment-free genome tree inference by learning group-specific distance metrics

Kaustubh R Patil et al. Genome Biol Evol. 2013.

Abstract

Understanding the evolutionary relationships between organisms is vital for their in-depth study. Gene-based methods are often used to infer such relationships, which are not without drawbacks. One can now attempt to use genome-scale information, because of the ever increasing number of genomes available. This opportunity also presents a challenge in terms of computational efficiency. Two fundamentally different methods are often employed for sequence comparisons, namely alignment-based and alignment-free methods. Alignment-free methods rely on the genome signature concept and provide a computationally efficient way that is also applicable to nonhomologous sequences. The genome signature contains evolutionary signal as it is more similar for closely related organisms than for distantly related ones. We used genome-scale sequence information to infer taxonomic distances between organisms without additional information such as gene annotations. We propose a method to improve genome tree inference by learning specific distance metrics over the genome signature for groups of organisms with similar phylogenetic, genomic, or ecological properties. Specifically, our method learns a Mahalanobis metric for a set of genomes and a reference taxonomy to guide the learning process. By applying this method to more than a thousand prokaryotic genomes, we showed that, indeed, better distance metrics could be learned for most of the 18 groups of organisms tested here. Once a group-specific metric is available, it can be used to estimate the taxonomic distances for other sequenced organisms from the group. This study also presents a large scale comparison between 10 methods--9 alignment-free and 1 alignment-based.

Keywords: alignment; alignment-free; distance metric learning; genome signature; genome tree; sequence comparison; taxonomy.

PubMed Disclaimer

Figures

F<sc>ig</sc>. 1.—
Fig. 1.—
Performance on the phylogenetic groups. Each bar shows a performance measure along with error bars showing SD.
F<sc>ig</sc>. 2.—
Fig. 2.—
Performance on the GC-content groups. Each bar shows a performance measure along with error bars showing SD.
F<sc>ig</sc>. 3.—
Fig. 3.—
Performance on the ecological groups from three attributes. The bars show the performance measures and the error bars indicate SD.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–410. - PubMed
    1. Blaisdell BE. A measure of the similarity of sets of sequences not requiring sequence alignment. Proc Natl Acad Sci U S A. 1986;83:5155–5159. - PMC - PubMed
    1. Burge C, Campbell AM, Karlin S. Over- and under-representation of short oligonucleotides in DNA sequences. Proc Natl Acad Sci U S A. 1992;89:1358–1362. - PMC - PubMed
    1. Ciccarelli FD, et al. Toward automatic reconstruction of a highly resolved tree of life. Science. 2006;311:1283–1287. - PubMed
    1. Coenye T, Gevers D, Van de Peer Y, Vandamme P, Swings J. Towards a prokaryotic genomic taxonomy. FEMS Microbiol Rev. 2005;29:147–167. - PubMed

Publication types

LinkOut - more resources

-