A novel method to analyse DART TOFMS spectra based on Convolutional Neural Networks: A case study on methanol extracts of wool fibres from endangered camelids

International Journal of Mass Spectrometry, vol. 489, pp. 117050

Abstract

Monitoring the illegal trade of wool fibres of wild vicuña (Vicugna vicugna) and guanaco (Lama guanicoe) is highly desirable. The high market value of fleece from these camelid species poses a threat to their wild populations. A previous study showed that direct analysis in real time time-of-flight mass spectrometry (DART-TOFMS) effectively identifies wool fibres to species. Producing high-resolution data in a short period of time makes DART-TOFMS a reliable identification tool, even though data analysis can still be improved. The present study proposes a novel data analysing pipeline based on Convolutional Neural Networks (CNN), applicable to any kind of DART-TOF MS data. We tested our proposed method on keratin fibres of four camelid species (Vicugna vicugna: n = 19; Vicugna pacos: n = 20; Lama guanicoe: n = 20, and Lama glama: n = 20). Analyses showed that selecting 512 ions with the highest relative intensity provides the best resolution and yields 100% accuracy for species identification.

Authors 9

1. Mehrdad Jahanbanifard corresponding Aachen Leiden Institute of Advanced Computer Science Naturalis Biodiversity Center

   Leiden University · Naturalis Biodiversity Center

   Affiliation as printed

   Leiden Institute of Advanced Computer Science (LIACS), Leiden University, Niels Bohrweg 1, 2333, CA, Leiden, the Netherlands

   Naturalis Biodiversity Center, Darwinweg 2, 2333, CR, Leiden, the Netherlands

2. Erin R. Price

   US Forest Service

   Affiliation as printed

   USFS International Programs Wood Identification and Screening Center, 1490 E. Main St., Ashland, OR, 97520, USA

3. Benito A. González

   University of Chile

   Affiliation as printed

   Faculty of Forest Sciences and Nature Conservation, Universidad de Chile, Chile

4. Luis Alberto Raggi

   University of Chile

   Affiliation as printed

   Faculty of Veterinary Sciences, Animal Sciences Department, Universidad de Chile, Chile

5. Shima Javanmardi Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science (LIACS), Leiden University, Niels Bohrweg 1, 2333, CA, Leiden, the Netherlands

6. Frederic Lens Aachen Institute of Biology Leiden Naturalis Biodiversity Center

   Leiden University · Naturalis Biodiversity Center

   Affiliation as printed

   Leiden University, Institute of Biology Leiden, Plant Sciences, Sylviusweg 72, 2333, BE, Leiden, the Netherlands

   Naturalis Biodiversity Center, Darwinweg 2, 2333, CR, Leiden, the Netherlands

7. Barbara Gravendeel Aachen Naturalis Biodiversity Center

   Naturalis Biodiversity Center · Radboud University Nijmegen

   Affiliation as printed

   Naturalis Biodiversity Center, Darwinweg 2, 2333, CR, Leiden, the Netherlands

   Radboud Institute for Biological and Environmental Sciences, Heyendaalseweg 135, 6500 GL Nijmegen, The Netherlands

8. Edgard O. Espinoza
   Affiliation as printed

   National Fish and Wildlife Forensic Laboratory, 1490 E. Main St., Ashland, OR, 97520, USA

9. Fons J. Verbeek Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science (LIACS), Leiden University, Niels Bohrweg 1, 2333, CA, Leiden, the Netherlands