More affordable and effective noninvasive single nucleotide polymorphism genotyping using high‐throughput amplicon sequencing

Abstract

Noninvasive genotyping methods have become key elements of wildlife research over the last two decades, but their widespread adoption is limited by high costs, low success rates and high error rates. The information lost when genotyping success is low may lead to decreased precision in animal population densities, which could misguide conservation and management actions. Single nucleotide polymorphisms (SNPs) provide a promising alternative to traditionally used microsatellites as SNPs allow amplification of shorter DNA fragments, are less prone to genotyping errors and produce results that are easily shared among laboratories. Here, we outline a detailed protocol for cost-effective and accurate noninvasive SNP genotyping using multiplexed amplicon sequencing optimized for degraded DNA. We validated this method for individual identification by genotyping 216 scats, 18 hairs and 15 tissues from coyotes (Canis latrans) using 26 SNPs. Our genotyping success rate for scat samples was 93%, and 100% for hair and tissue, representing a substantial increase compared to previous microsatellite-based studies while remaining at a low cost of under $5 per PCR replicate (excluding labour). The accuracy of the genotypes was further corroborated in that genotypes from scats matching known, GPS-collared coyotes were always located within the territory of the known individual. We also show that different levels of multiplexing produced similar results, but that PCR product cleanup strategies can have substantial effects on genotyping success. By making noninvasive genotyping more affordable, accurate and efficient, this research may allow for a substantial increase in the use of noninvasive methods to monitor and conserve free-ranging wildlife populations.

Evaluating and integrating spatial capture–recapture models with data of variable individual identifiability

“…We used genotyping by amplicon sequencing (Eriksson et al 2020) to identify unique individuals with single-nucleotide polymorphisms (SNPs), a process that uses the power of highthroughput sequencers to yield high genotype success rates and low error rates. Coyote scats were genotyped following methods in Eriksson et al (2020).…”

“…We used genotyping by amplicon sequencing (Eriksson et al 2020) to identify unique individuals with single-nucleotide polymorphisms (SNPs), a process that uses the power of highthroughput sequencers to yield high genotype success rates and low error rates. Coyote scats were genotyped following methods in Eriksson et al (2020). We genotyped bobcat, cougar, and black bears scat using the same approach but with novel primers designed and tested specifically for this study (Appendix S2).…”

“…We genotyped bobcat, cougar, and black bears scat using the same approach but with novel primers designed and tested specifically for this study (Appendix S2). Prior to genotyping, the identity of the defecator for each scat was identified using DNA metabarcoding of ~100 base pairs of the mitochondrial 12S gene region as part of a separate diet study (primers used in Eriksson et al 2020, Massey et al 2020, and modified from Riaz et al 2011). See Appendix S2 for SNP discovery methods and Eriksson et al (2020) for detailed primer design and genotyping protocols.…”

“…This would allow partial genotypes (Augustine et al 2019) or photos in which the marked status or individual identity (Augustine et al 2018, Jiménez et al 2019, Tourani et al 2020) cannot be ascertained to be included in SCR estimation. However, this model extension would minimally influence our results because of the high genotyping success from scats using genotyping by amplicon sequencing (Eriksson et al 2020) and our high success in identifying marked status or individual identity in our camera trap photos. Further, if the discrepancies between model types were driven by violations of model assumptions (e.g.…”

The Rapid Rise of Next-Generation Natural History

“…Spatial capture-recapture is the most data-hungry of these approaches because it requires the identification of unique individuals such as through camera trapping of uniquely marked individuals (e.g., jaguars, Sollmann et al, 2011) or with noninvasive genetic sampling (Sollmann et al, 2013;Augustine et al, 2019). As the efficiency of fecal genotyping has increased with improved molecular approaches (Eriksson et al, 2020) and the use of scat detection dogs, spatial capture-recapture has become the canonical statistical framework for estimating population density when animal identity is known. Genetic data can also be incorporated using close-kin mark recapture (first proposed by Skaug, 2001) based on the principle that an individual's genotype can be considered a recapture of its parents' genotypes, which allows abundance to be estimated based on the number of parentoffspring pairs in a sample.…”