Skip to main content
ARS Home » Plains Area » Manhattan, Kansas » Center for Grain and Animal Health Research » Stored Product Insect and Engineering Research » Research » Publications at this Location » Publication #309049

Research Project: PRESERVATION, ENHANCEMENT, AND MEASUREMENT OF GRAIN QUALITY AND MARKETABILITY

Location: Stored Product Insect and Engineering Research

Title: A near-infrared spectroscopy routine for unambiguous identification of cryptic ant species

Author
item Kinzer, Martin-carl - University Of Innsbruck
item Wagner, Herbert - University Of Innsbruck
item Peskoller, Andrea - University Of Innsbruck
item Moder, Karl - University Of Natural Resources & Applied Life Sciences - Austria
item Dowell, Floyd
item Arthofer, Wolfgang - University Of Innsbruck
item Schlick-steiner, Birgit - University Of Innsbruck
item Steiner, Florian - University Of Innsbruck

Submitted to: PeerJ
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/11/2015
Publication Date: 9/15/2015
Citation: Kinzer, M., Wagner, H.C., Peskoller, A., Moder, K., Dowell, F.E., Arthofer, W., Schlick-Steiner, B.C., Steiner, F.M. 2015. A near-infrared spectroscopy routine for unambiguous identification of cryptic ant species. PeerJ. 3:e991. doi: 10.7717/peerj.991.

Interpretive Summary: The identification of insect species is not always straightforward as similar species present a hurdle for traditional species discrimination. Fibre-optic near-infrared spectroscopy (NIRS) is a rapid and cheap method for a wide range of different applications, among them the identification of species. Despite its efficiency, NIRS has never been tested on a group of more than two species, and a working routine is still missing. Hence, we tested if specimens of the four morphologically highly similar, but genetically distinct ant species can be identified using NIRS. Furthermore, we evaluated which of three analysis tools was most efficient in species identification. Our NIRS identification routine with partial least squares regression was successful with up to 80% of identified specimens correctly classified. We emphasise our classification routine using fibre-optic NIRS was a highly efficient pre-screening identification method for similar ant species.

Technical Abstract: The identification of species – of importance for most biological disciplines – is not always straightforward as cryptic species present a hurdle for traditional species discrimination. Fibre-optic near-infrared spectroscopy (NIRS) is a rapid and cheap method for a wide range of different applications, among them the identification of species. Despite its efficiency, NIRS has never been tested on a group of more than two cryptic species, and a working routine is still missing. Hence, we tested if specimens of the four morphologically highly similar, but genetically distinct ant species Tetramorium alpestre, T. caespitum, T. impurum, and T. sp. B, all four co-occurring above 1300 m a.s.l. in the Alps, can be unambiguously identified using NIRS. Furthermore, we evaluated which of the three analysis tools, partial least squares regression (PLS), artificial neural networks (ANN), and random forests (RF), is most efficient in species identification. We opted for a 100% classification certainty, i.e. a residual risk of misidentification of zero within the available data, at the cost of excluding specimens from identification. Additionally, we examined which strategy, one-vs-all, i.e. one species compared to the pooled set of the remaining species, or binary-decision strategies, is best to reduce a multi-class system to a 2-class system, as is necessary for PLS. Our NIRS identification routine, based on a 100% identification certainty, was successful with up to 80% of unambiguously identified specimens of a species. In detail, PLS scored best over all species (43.3% of specimens), while RF was much less effective (8.3%) and ANN failed completely (0%). Moreover, we showed that the one-vs-all strategy is the only acceptable option to reduce multi-class systems because of a minimum expenditure of time. We emphasise our classification routine using fibre-optic NIRS in combination with PLS and the one-vs-all strategy as a highly efficient pre-screening identification method for cryptic ant species and possibly beyond.