Molecular identification of birds: performance of distance-based DNA barcoding in three genes to delimit parapatric species.

Neil John Gemmell,Mansour Aliabadian,Vincent Nijman,Mohammad Kaboli,Miguel Vences

doi:10.1371/journal.pone.0004119

Abstract

BackgroundDNA barcoding based on the mitochondrial cytochrome oxidase subunit I gene (cox1 or COI) has been successful in species identification across a wide array of taxa but in some cases failed to delimit the species boundaries of closely allied allopatric species or of hybridising sister species.Methodology/Principal FindingsIn this study we extend the sample size of prior studies in birds for cox1 (2776 sequences, 756 species) and target especially species that are known to occur parapatrically, and/or are known to hybridise, on a Holarctic scale. In order to obtain a larger set of taxa (altogether 2719 species), we include also DNA sequences of two other mitochondrial genes: cytochrome b (cob) (4614 sequences, 2087 species) and 16S (708 sequences, 498 species). Our results confirm the existence of a wide gap between intra- and interspecies divergences for both cox1 and cob, and indicate that distance-based DNA barcoding provides sufficient information to identify and delineate bird species in 98% of all possible pairwise comparisons. This DNA barcoding gap was not statistically influenced by the number of individuals sequenced per species. However, most of the hybridising parapatric species pairs have average divergences intermediate between intraspecific and interspecific distances for both cox1 and cob.Conclusions/SignificanceDNA barcoding, if used as a tool for species discovery, would thus fail to identify hybridising parapatric species pairs. However, most of them can probably still assigned to known species by character-based approaches, although development of complementary nuclear markers will be necessary to account for mitochondrial introgression in hybridising species.

Highlights

Mitochondrial DNA markers have been widely applied in molecular phylogenetic studies, but deciding which mtDNA genes to use for the identification of species remains an important issue [1,2,3] because different parts of the mtDNA genome evolve at different mutation rates [4,5]
MtDNA, with rapid pace of sequence changes, regularly shows pronounced divergences between closely related species [1,6] but concern has been expressed that mtDNA sequence differences among such closely related species will often be too small to allow their discrimination, and the problem will be further accentuated by phenomena of ancient sharing of haplotype polymorphisms and by introgression (e.g. [7])
Recent studies suggest that sequences of the mitochondrial cytochrome oxidase subunit I could serve as a fast and accurate marker for the identification of animal species, and for the discovery of new species across the tree of life [8], a procedure for which the term DNA barcoding has been coined

Summary

Introduction

Mitochondrial DNA (mtDNA) markers have been widely applied in molecular phylogenetic studies, but deciding which mtDNA genes to use for the identification of species remains an important issue [1,2,3] because different parts of the mtDNA genome evolve at different mutation rates [4,5]. Recent studies suggest that sequences of the mitochondrial cytochrome oxidase subunit I (cox or COI) could serve as a fast and accurate marker for the identification of animal species, and for the discovery of new species across the tree of life [8], a procedure for which the term DNA barcoding has been coined. Variation of cox sequences within species was an average of 20 times smaller than between species, and there was a clear gap between intra- and interspecific variation Utilizing this barcoding gap, a standard sequence threshold was proposed to define species boundaries of around 10 times the mean intraspecific variation for the group under study [9]. DNA barcoding based on the mitochondrial cytochrome oxidase subunit I gene (cox or COI) has been successful in species identification across a wide array of taxa but in some cases failed to delimit the species boundaries of closely allied allopatric species or of hybridising sister species

Methods

Results

Conclusion