A probabilistic model for designing and assessing the performance of eDNA sampling protocols. (29th November 2019)

• Record Type: Journal Article
• Authors: Song, Jeffery W.; Schultz, Martin T.; Casman, Elizabeth A.; Bockrath, Katherine D.; Mize, Erica; Monroe, Emy M.; Tuttle‐Lau, Maren; Small, Mitchell J.
• Is Part Of: Molecular ecology resources. Volume 20:Number 2(2020:Mar.)
• Page Start: 404
• Page End: 414
• Publication Date: 2019-11-29
• Subjects: detection sensitivity -- environmental DNA -- sampling protocol; Molecular ecology -- Periodicals; 572.8
• DOI: 10.1111/1755-0998.13113
• Languages: English
• ISSNs: 1755-098X

Abstract

Abstract: Environmental DNA (eDNA) sampling, the detection of species‐specific genetic material in water samples, is an emerging tool for monitoring aquatic invasive species. Optimizing eDNA sampling protocols can be challenging because there is imperfect understanding of how each step of the protocol influences its sensitivity. This paper develops a probabilistic model that characterizes each step of an eDNA sampling protocol to evaluate the protocol's overall detection sensitivity for one sample. The model is then applied to analyse how changes over time made to the eDNA sampling protocol to detect bighead (BH) and silver carp (SC) eDNA have influenced its sensitivity, and hence interpretation of the results. The model shows that changes to the protocol have caused the sensitivity of the protocol to fluctuate. A more efficient extraction method in 2013, new species‐specific markers with a qPCR assay in 2014, and a more efficient capture method in 2015 have improved the sensitivity, while switching to a larger elution volume in 2013 and a smaller sample volume in 2015 have reduced the sensitivity. Overall, the sensitivity of the current protocol is higher for BH eDNA detection and SC eDNA detection compared to the original protocol used from 2009 to 2012. The paper shows how this model of eDNA sampling can be used to evaluate the effect of proposed changes in an eDNA sampling and analysis protocol on the sensitivity of that protocol to help researchers optimize their design.