Identifying individual ungulates from fecal DNA: a comparison of field collection methods to maximize efficiency, ease, and success

Abstract

Non-invasive genetic sampling can facilitate the identification of individual animals across a landscape, with applications to management and conservation. Fecal material is a readily available source of DNA, and various methods exist for collecting fecal samples for DNA preservation. In particular, swab methods offer considerable promise, but their utility in real-world field contexts remains relatively untested. We systematically compared multiple genetic fecal sampling methods across all stages of data collection and analysis, including sampling in the field, DNA extraction in the lab, and identification of individuals using microsatellite genotyping. We collected 112 fecal samples from black-tailed deer (Odocoileus hemionus columbianus) in the field in Mendocino County, California, across a range of sample conditions of unknown age. We systematically compared the efficiency, ease, and genotyping success of three methods for field collection and storage of ungulate fecal samples: whole pellets in ethanol, whole dry pellets in paper envelopes, and cotton swabs in buffer. Storage method, sample condition, and their interaction predicted genotyping success in the top binomial GLMMs. We found that swabbing pellets resulted in the greatest percentage of individually identifiable genotypes (81%, compared to 60% for dry samples and 56% for ethanol), despite lower DNA concentrations. While swabbing pellets requires a greater time investment in the field, the samples are easier and safer to store and transport, and subsequent labwork is more efficient as compared to whole-pellet collection methods. We, therefore, recommend the swab method for most contexts. We provide additional recommendations and field protocols based on subsequent collection of 2284 swab samples for a larger monitoring study of the deer population, given that this large number of samples spanned a range of sample conditions and time spent in storage.

A spatial, closed integrated population model to estimate wildlife population size and structure

“…To account for the potential of false alleles and allelic dropouts due to low-quality DNA, we repeated the genetic analysis of each sample twice and used the R package allelematch to assign individuals based on probabilities of allele mismatches (Galpern et al 2012). For a full description of the genetics methods that we used, see Bach et al (2022).…”

Capturing wild animal welfare: a physiological perspective

“…Without individual identification, only aggregate measures of welfare can be calculated at the population level (as frequently seen in epidemiological studies; Broome et al ., 2002). This problem of individual identification can, however, be resolved by employing molecular identification approaches to the collected biological material (Zemanova, 2021; Bach et al ., 2022).…”

An experimental design for obtaining DNA of a target species and its diet from a single non‐invasive genetic protocol

“…One of the protocols has mainly been used for studies on target species (e.g. Alberts et al., 2017 ; Angom et al., 2020 ; Bach et al., 2022 ; Biswas et al., 2019 ; Bourgeois et al., 2019 ; Modi et al., 2021 ; Natesh et al., 2019 ; Ramón‐Laca et al., 2015 ), while the other has mostly been used in dietary studies (e.g. Alberdi et al., 2018 ; Ando et al., 2020 ; Craine, 2021 ; Craine et al., 2015 , 2016 ; Jesmer et al., 2020 ; Jorns et al., 2020 ; Pansu et al., 2019 ; van Zinnicq Bergmann et al., 2021 ).…”