Noninvasive Genetic Sampling in Wildlife Management: The Past, Present, and Future

Symposium

SESSION NUMBER: 79

Symposia will be available on-demand on their scheduled date, then again at the conclusion of the conference.

 
The development of noninvasive genetic sampling in the 1990s revolutionized wildlife management and our ability to monitor wild populations. The advent of noninvasive genetic sampling ? extracting genetic material from environmental DNA sources such as hair, feces, or feathers ? has enabled researchers to gather critical data about wild populations without handling an animal. It has also allowed for increased sample sizes (e.g. it?s easier to pick up 100 feces than trap 50 wolves); provided necessary data to estimate population size; increased detection of cryptic species; and facilitated the assessment of population connectivity. Despite the revolution of noninvasive genetic sampling techniques, these methods are limited by the quality of environmental source DNA. Advances in sequencing technology are actively expanding the application of noninvasive genetic sampling. For instance, we can capture mosquitoes and flies, or collect soil and water samples to study wildlife in a region. We can also assess an individual?s genotype, diet, microbiome, and disease state with one fecal sample at increasingly reduced costs. These novel extensions of noninvasive genetic sampling are ushering in a new methodological era that will radically advance the fields of population biology and community ecology. This symposium will discuss past, present, and future applications of noninvasive genetic sampling, how these methods are expanding our sampling abilities, and the practical application of new techniques. Please join us for an engaging lineup of talks that will be accessible for anyone interested in how noninvasive genetic sampling can improve wildlife management, no genetic experience required!

Development of a Noninvasive SNP Assay to Monitor the Genomic Health of Newly Introduced Isle Royale Wolves
Samuel D. Hervey; Mark C. Romanski; Kristin E. Brzeski
Isolation from other mainland populations reduces genetic diversity, impacting evolutionary potential and making deleterious alleles associated with reduced fitness more apparent. This issue was well demonstrated over the last two decades for the gray wolves (Canis lupus) of Isle Royale. In 2017, only two native wolves remained on the island, which were presumed unlikely to produce viable offspring due to a high level of inbreeding. As a result, 20 gray wolves from Michipicoten, Ontario (n=8), Jostle Lake, Ontario (n=3), Grand Portage, Minnesota (n=5), and Baraga County, Michigan (n=4) were introduced to the island of Isle Royale, in 2018 and 2019. The reintroduction of gray wolves to Isle Royale presents the opportunity to study the population at a genomic-scale, providing novel insights into the health of the population. Preliminary results generated from microsatellite markers demonstrate Michipicoten wolves relocated to Isle Royale are highly related. However, overall genetic diversity is high, and all relocated wolves have a high probability of assignment to gray wolf relative to eastern wolf (Canis lupus lycaon). To inform future management decisions we are evaluating baseline genomic variation of the translocated wolves and predicting the sustainability of genetic variation through simulated mating scenarios, but importantly, we need a robust noninvasive genetic monitoring program to manage population health over the next decade. This talk will focus on our strategy to develop a candidate list of single nucleotide polymorphisms (SNPs) to use for long-term noninvasive sampling. Utilizing high quality DNA samples obtained during wolf translocation, we generated a candidate list of SNPs using genotyping-in-thousands by sequencing (GT-seq) to genotype noninvasive scat samples and optimize an Isle Royale specific genomic monitoring program. This assay will be critical for pedigree construction and assessment of inbreeding through time, ideally helping prevent another population collapse.
Noninvasive Genetics: What We Have Learned and What Is Next for Wildlife Management
Jennifer R. Adams; Lisette P. Waits
Non-invasive genetic sampling (gNIS) and subsequent DNA analysis of wildlife populations can be described as the collection and analysis of wildlife DNA left behind in the environment and is an important tool used in the field of conservation genetics. The first studies using this technique appeared in the literature in the early 1990s and thirty years later, gNIS has gained acceptance within the wildlife management profession. GNIS offered researchers the ability to sample larger areas and more individuals than traditional survey methods which proved useful when working with wide ranging, rare and/or difficult to capture species. One challenge with gNIS has always been the low quantity and degraded quality of the DNA obtained from non-invasive sample types like hair and feces. In the beginning, this limitation meant the first studies were costly, time consuming and suffered from low sample success rates and genotyping errors. This problem delayed the use of gNIS methods for many species and often restricted the application of gNIS to endangered populations or well funded research and management projects. Advancing technologies and methodologies have found ways to mitigate these limitations, and gNIS can now be cost-effectively used to answer questions about occupancy, population abundance, individual relatedness, hybridization, disease, predation, diet, pregnancy rates and population substructure with increasingly indirect sample types like water from lakes and rivers to soil. Cutting edge sequencing technologies have unlocked the ability to delve further into research questions pertaining to adaptation and evolution and are expanding applications of gNIS for wildlife management and conservation as we transition into the next era of genetic and genomic analyses.
Optimizing noninvasive genetic sampling: considerations for long-term monitoring
Robert C. Lonsinger; Lisette P. Waits
Noninvasive genetic sampling (NGS) can offer a cost-effective and reliable way to conduct long-term monitoring of wildlife populations. Although the costs of genetic procedures have decreased, costs may still limit NGS-based monitoring when working at large spatial scales or when sympatric species are indistinguishable from hair or scat. We used kit fox (Vulpes macrotis) data generated with NGS (via scats) as a model system to demonstrate how optimization techniques can reduce costs, while maintaining data reliability. We focus on data collected for estimating abundance and occupancy. We present a conceptual model for combining sample accumulation and DNA degradation rates to identify the most efficient temporal sampling design for capture-recapture analyses. We then applied this model and compared the performance of three abundance estimators. Two estimators (Huggins and spatially-explicit capture-recapture) employed temporal replication. In contrast, one estimator (capture with replacement; CAPWIRE) required only one sampling occasion and potentially offered a low-cost alternative. Finally, we investigated how sample identification criteria and data handling procedures influenced occupancy analyses. Cost-efficiency of NGS was improved with a temporal design that balanced field and laboratory costs. Choice of abundance estimator and sampling design significantly influenced estimates: Huggins and spatially-explicit models produced similar estimates, but estimates from CAPWIRE tended to be biased low with high precision. For occupancy, alternative sample identification criteria (without laboratory costs; i.e., field-based and statistical classification tree-based) produced occupancy estimates that were biased high with poor precision, as did a laboratory-based removal design that had only limited reductions in costs. In contrast, combining classification tree and genetic identification, or using a within-survey conditional-replicate design, produced occupancy estimates that were similar to ‘truth’ at savings of up to 88%. Our results demonstrate how pilot data can be used to optimize NGS and we provide recommendations for the application of NGS for long-term monitoring.
Estimating Nuclear Genetic Diversity from Environmental DNA Samples: A Proof of Concept with Hellbender eDNA
Stephen Spear; Austin Russell; JJ Apodaca
Environmental DNA has become a valuable approach to detect presence of a variety of secretive species, especially those that occur in aquatic environments. The method also holds promise for monitoring population status and abundance. A common approach has been to use concentration or amount of eDNA present in a sample to infer population size. While this method has worked in some cases, for many species there is a not a strong correlation between eDNA amount and number of individuals or biomass. The Eastern Hellbender has established protocols for eDNA detection, but attempts to estimate size or density through eDNA amount have been equivocal at best. Another avenue for estimating population status would be to estimate genetic diversity at variable markers. Microsatellite DNA represents one such genetic marker as microsatellite loci have been developed for hellbenders and data exist on rangewide genetic diversity at these markers. However, amplifying nuclear loci such as microsatellites can be difficult for degraded samples such as eDNA due to the lower amount of nuclear DNA in a sample compared to mitochondrial DNA. We developed and optimized a protocol to amplify hellbender microsatellite loci using a two-phase nested PCR approach, along with a DNA purification and concentration step. We further tested two approaches to genotyping these loci: 1) fragment analysis that is the traditional genotyping for microsatellites and 2)sequencing microsatellite loci using next-generation sequencing platforms. Our results demonstrate the ability to amplify microsatellite loci at a variety of eDNA concentrations, but also show that allelic dropout is a significant source of error at lower concentrations. Therefore, future directions would be to determine the minimum eDNA concentration for reliable genotyping as well as to develop models to translate microsatellite allelic diversity to estimates of population size.
Exploring Coyote Scavenging Ecology using Noninvasive Techniques in a Hyper-abundant Elk System
Charlotte Eriksson; Joel Ruprecht; Taal Levi
The study of trophic interactions and energy transfer between trophic levels is a fundamental aspect of ecology. Predator-prey interactions often receive the most attention, but scavenging plays an essential role in structuring and stabilizing food-webs as well. Previous scavenging studies have relied on direct observation, camera traps, or inspection of carcasses to draw conclusions about scavenging ecology. Importantly, these methods rely on locating carcasses via snow-tracking carnivores, telemetered individuals, or through experimental placement of carcasses on the landscape. An unexplored method is to genotype both the scavenger and carrion species from DNA in carnivore scats. This method is entirely noninvasive, and avoids the financial, logistical, and animal welfare burdens of deploying telemetry devices to identify scavengers or to locate carcasses. In this study we genotyped both coyotes and elk found in coyote scats that were collected by scat-detection dog teams in northeastern Oregon. We first confirmed that the scats were from coyotes and that they contained elk by DNA metabarcoding using universal vertebrate primers. We then identified the individual coyote and elk using single-nucleotide polymorphisms (SNPs) and high-throughput amplicon sequencing. Using these techniques and GPS data from a concurrent study of collared coyotes we describe coyote diet diversity and scavenging behavior on a population and individual level.
Transitioning from Microsatellite Loci to Single Nucleotide Polymorphism (SNP) Loci for Fecal DNA Genetic Monitoring of the Endangered Columbia Basin Pygmy Rabbit (Brachylagus idahoensis).
Lisette Waits; Stacey Nerkowski; Paul Hohenlohe
Loss and fragmentation of habitat has led to the near extirpation of the disjunct pygmy rabbit population in the Columbia Basin of Washington State. In 2003, the CB pygmy rabbit (CBPYRA) was listed as an endangered distinct population segment under the US Endangered Species Act. In 2001, sixteen CBPYRA were taken from the last remaining population in Washington to start a captive breeding program, and Idaho rabbits were added to counteract the negative effects of inbreeding. Rabbits were moved to semi-wild breeding enclosures at the recovery site in Washington in 2011, and additional rabbits were translocated from other populations within the western United States in 2011 and 2020 to bolster the size of the captive population. Since the initiation of onsite captive breeding, ~2009 mixed ancestry rabbits have been released into the wild, and the primary monitoring method has been winter fecal pellet collection and analysis using 19 nuclear DNA microsatellite loci. We are have recently identified over 14,000 SNP loci for pygmy rabbits and have designed a 500 locus SNP panel for GTseq genotyping that we can use for analysis of individual identity, parentage, ancestry, and adaptive differences. This panel will allow us to genetically monitor captive and wild populations of CBPYRA to guide strategies for conservation and management. This presentation will provide an overview of our GTseq analysis methods, qPCR assay development, genotyping success rates from different DNA sources, costs, as well as the benefits and challenges of transitioning from a microsatellite-based genetic monitoring program to SNP-based genetic monitoring program.

 
Organizers: Kristin Brzeski: Michigan Technological University, Houghton, MI; Caitlin Ott-Conn: Michigan Department of Natural Resources, Lansing, MI; Robert Lonsinger, South Dakota State University, Brookings, SD; Stephen Spear, The Wilds, Cumberland, OH
 
Supported by: Molecular Ecology Working Group

Symposium
Location: Virtual Date: September 30, 2020 Time: -