To Go Boldly … : The Genomics Frontier of Conservation and Management: Part I


8:00AM Developing a Dual-Purpose Gtseq Panel for Stock Discrimination and Parentage Analysis in Walleye
  Kristen Gruenthal
Previous genetic research on walleye (Sander vitreus) has provided vital information used toward stewardship of this culturally and recreationally important species. However, currently available genetic tools, such as microsatellites, do not provide enough power for assignment of parentage or stock discrimination. We leveraged RADseq data to identify tens of thousands of high-resolution single nucleotide polymorphisms (SNPs) and designed a pilot GTseq panel incorporating 600 loci, covering nearly 1800 SNPs, based on microhaplotype heterozygosity and utility for genetic differentiation. We will discuss the trade-offs we experienced between choosing loci for population discrimination and parentage, as well as provide a roadmap for developing GTseq panels useful for multiple applications. We will also discuss our workflow for low-cost high-throughput genotyping. Finally, our walleye panel represents a powerful new tool that will provide important data for researchers and natural resource agencies across the Upper Midwest, and we outline some of its potential utility for fisheries management.
8:20AM Comparing Edna with Conventional Methods for Bio-Monitoring of Fish Communities
  Louis Bernatchez
Accurate data on distribution and abundance are critical for conservation and management of biodiversity. In aquatic ecosystems, several inventory methods are widely used to estimate those parameters. Despite their positive assets, these methods are not without constraints and limitations. For instance, they can be invasive, costly in terms of material and human resources, may cause unwanted mortality in communities studied as well as being subject to size and species selection bias. Others such as hydro-acoustic methods may leave uncertainties regarding species composition. Environmental DNA (eDNA) analysis could be used as a non-invasive, less costly, and perhaps more accurate alternative or in complement to conventional bio-monitoring methods. Yet, eDNA methods potentially come with their own caveats and therefore, there is a need for rigorous comparisons with conventional inventory methods. We will evaluate the pros and cons between eDNA (both single species and eMetabarcoding approaches) and three conventional methods (gillnet, trawler, hydro-acoustic) for monitoring freshwater and marine fish communities in terms of species identification, richness, relative abundance and species abundance. The results will be discussed in the context of their applications for bio-monitoring in different contexts including environmental assessment, commercial fisheries and whale conservation.
8:40AM Fast and Easy: CRISPR-Based Assays for Rapid and Accurate Species Genetic Identification
  Melinda Baerwald
CRISPR-based technology is revolutionizing the healthcare industry and other scientific sectors but has seen limited use in molecular ecology to date. We have developed CRISPR assays capable of distinguishing single base pair changes between nucleic acid templates with remarkable sensitivity and accuracy. Additionally, this tool is rapid and does not require initial DNA extraction to produce reliable molecular diagnostic results. As a proof of principle, we created assays that genetically differentiate three closely related fish species co-occurring in the San Francisco Estuary: the federal and state listed Delta Smelt, the state listed Longfin Smelt, and the non-native Wakasagi. Work is in progress to convert the assay to a field-deployable method to enable real-time species or subspecies diagnostics in any location. We believe this technology has the potential to transform how genetic identification is performed across the field of molecular ecology to aid in efficient and accurate species monitoring and management.
9:00AM Mechanisms and Consequences of Seasonal and Interannual Distribution Shifts of Pacific Cod from Pooled Whole Genome Sequences
  Carolyn Tarpey
Changing oceanic conditions in the North Pacific have resulted in unprecedented shifts of the distribution of marine species in recent years. In Pacific cod, changing environmental conditions may affect both seasonal migration between spawning and feeding locations as well as the long term spatial distribution of abundance. In this study, we investigate the mechanisms and consequences of such distribution shifts by pooled whole genome sequencing of twelve different spawning aggregations from the Aleutian Islands, Bering Sea and Gulf of Alaska. To analyze this very large data set, we used the Broad Institute’s bioinformatics platform Terra to run pipelines on Google Cloud which allows access to nearly unlimited computational resources. We processed data by aligning sequences to the Atlantic cod genome and called variant genotypes using the program GATK4, resulting in hundreds of thousands of SNPs across the Pacific cod genome. These SNPs are used to estimate genetic differentiation across the genome, thus identifying adaptive differentiation as well as neutral population structure. Our results will be invaluable for sustainable management of an important fishery in a rapidly changing environment.
9:20AM Is the Red Wolf a Hybrid Species? Revisiting Genomic Data Using Coalescent Predictions
  Tyler Chafin
The red wolf (Canis rufus) is hypothesized to originate from hybridization between North American gray wolves (Canis lupus) and coyotes (Canis latrans). However, alternate scenarios have been proposed, including possible ecotypy of red wolves as a subset of the gray wolf, with subsequent hybridization caused by the contemporary range expansion of coyotes in the Anthropocene. We here present an attempt to resolve the evolutionary history of Canis rufus by re-visiting publicly available whole genome data within a novel hypothesis testing framework. We generate explicit predictions of the genome-wide coalescent time frequency distribution for a three-taxon case under multiple alternative treelike and non-treelike evolutionary scenarios, thereby allowing discrimination of homoploid hybrid speciation from post-isolation introgression. We also validate these predictions using a coalescent hidden Markov model which models shifts in topological and coalescent histories along chromosomes. We found support for contemporary gene flow from coyotes following ancient divergence of Canis rufus from a shared common ancestor with Canis lupus, suggesting that evolutionary independence of C. rufus pre-dates admixture with coyote. Genome blocks in the red wolf genome derived from coyote were dominant (e.g. comprising 2/3 of chromosome 1), and were significantly larger (m=17k), suggesting a relatively recent exchange.
09:40AM Break
1:10PM The Future Is Now: Amplicon Sequencing and Sequence Capture Usher in the Conservation Genomics Era
  Wesley Larson
The genomics revolution has initiated a new era of population genetics where genome-wide data are frequently used to understand complex patterns of population structure and selection. However, the application of genomic tools to inform management and conservation has been somewhat rare outside a few well-studied species. Fortunately, two recently developed approaches, amplicon sequencing and sequence capture, have the potential to significantly advance the field of conservation genomics. Both approaches allow sequencing of thousands of individuals at relatively low costs, do not require any specialized equipment for library preparation, and generate data that can be analyzed without sophisticated computational infrastructure. In this talk, we discuss the advantages and disadvantages of each method and provide a decision framework for geneticists who are looking to integrate these methods into their research program. Both of these techniques should smooth the transition from traditional genetic techniques to genomics, helping to usher in the conservation genomics era.
1:30PM Low-Coverage Whole Genome Sequencing Analysis of Atlantic Cod Stock Structure Reveals Insights for Fishery Management
  Adrienne Kovach
Low coverage whole genome sequencing enables scaling up sample sizes in studies of natural populations to maximize the breadth of genomic data collected from hundreds of individuals needed to address conservation and management issues. To demonstrate the utility of this approach, we present an analysis of stock structure in Atlantic cod of 300 individuals from 20 populations in U.S. and adjacent Canadian waters. Using genotype likelihoods estimated at 11 million SNPs, we found haplotype divergence in known chromosomal inversions among Canadian and U.S. sampling locations. We identified novel FST outlier peaks that differentiated sampling locations. Comparisons between allochronic winter and spring spawning populations revealed outlier peaks near genes involved in reproduction, particularly genes associated with the brain-pituitary-gonadal axis, which may shed light on the mechanism controlling timing of spawning. We also found outlier peaks associated with heat shock proteins and other genes of functional relevance, suggesting multifaceted selection pressures and local adaptation among spawning populations. Our results provide a high-resolution picture of Atlantic cod population structure that reveals greater biocomplexity than is recognized in current management structures. More broadly, we highlight the application of low coverage whole genome sequencing for considering evolutionary potential in fisheries and wildlife management decisions.
1:50PM Genomic Changes across the Evolutionary History of the VHSV-IVb Fish Virus
  Carol Stepien
Viral Hemorrhagic Septicemia virus (VHS) first appeared in the Laurentian Great Lakes in the early 2000s as a new and novel substrain (IVb), killing >30 fish species in large outbreaks from 2005-2010. A long interlude followed, which preceded two small, allopatric, and constrained 2017 outbreaks; none have occurred since. We employ a genomics, phylogenetic, and population genetics approach to evaluate the virus’ evolutionary trajectory, analyzing G-gene sequences from 185 isolates (32 haplotypes) and 46 entire genomes. Its genomes had 253 substitutions (2.3% of all nucleotides); 85 (16.6%) were nonsynonymous. The greatest proportion occurred in the non-coding region (ncds; 4.3%) followed by the Nv (3.8%) and M genes (2.8%). The most amino acid changes were in the M gene (52.9%), followed by Nv (50.0%), G (48.6%), N (35.7%) and L (23.1%). Population genetic analyses revealed pronounced spatial genetic divergence among the Upper, Central, and Lower Great Lakes, along with significant differentiation and diversification over time. VHSv-IVb thus markedly evolved across its
2:10PM Elucidating Plague Transmission Among Prairie Dog Colonies By Using Vector and Host Genomics
  Rachael Giglio
Plague is a flea-mediated disease caused by the bacterium Yersinia pestis. It moves through prairie dog colonies quickly, causing >90% mortality. However, the transmission pathways among prairie dog colonies remain unclear. Prairie dogs live in highly structured social groups with limited gene flow among colonies. Yet, we found that the main flea species found on black-tailed prairie dogs, Oropsylla hirsuta, has a higher degree of gene flow among prairie dog colonies than the prairie dogs themselves. This suggests that an alternative host may serve to move fleas among colonies. To evaluate the potential for small rodents to carry plague-infected fleas among prairie dog colonies, we used cutting-edge genomic techniques to characterize patterns of gene flow for small rodent species and the flea vector most prominently found on prairie dog colonies (O. hirsuta). We found that patterns of gene flow in prairie dog fleas exhibit greater concordance with deer mice, Peromyscus maniculatus, than with black-tailed prairie dogs. This means that deer mice may be a viable alternative host to move prairie dog fleas among colonies, exacerbating the spread of plague.
2:30PM Genomic Analysis of Population History and Reproduction for Hawaiian Monk Seals
  Samantha Hauser
The Hawaiian monk seal, one of the most critically endangered marine mammals, has recently stabilized after years of intensive conservation efforts. Here we use genomic technologies to address two key gaps in our understanding of localized population trends: the legacy of a recent localized expansion and variance in reproductive success. Populations in the Main Hawaiian Islands have been increasing since the 1980s and recently stabilized, as such we were interested in the legacy of migrants from the Northwest Hawaiian islands in expanding populations. We used BestRAD sequencing to genotype 192 individuals from across the 10 islands to investigate the source of the recent population expansion. We detected genomic signatures of a population expansion in the Main Hawaiian Islands and evaluated alternative immigration hypotheses. Secondly, we used pedigree-based analysis of a local population to estimate the variance in reproductive success among individuals, a key parameter for predicting future population trends. A genomics approach was critical in this case, to quantify genetic diversity and structure in populations with low genetic variation, provide accurate estimates of key population parameters, and interpret complex population histories. Recovery of the endangered Hawaiian monk seal will benefit from the incorporation of genomic data into management plans.
2:50PM Refreshment Break
3:20PM Using Genomic Tools to Investigate Adaptive Diversity in Great Lakes Cisco
  Amanda Ackiss
The Great Lakes were once populated by a diverse Coregonus artedi (cisco) species complex comprised of at least eight distinct morphotypes. Over the past century, the introduction of invasive forage fish, overfishing, and habitat loss led to large decreases in cisco abundance and lake-wide extirpation to complete extinction of historically documented forms. Recent evidence for declining abundance of invasive fish and increasing abundance in cisco has led to growing interest in re-establishing lost populations. An understanding of phenotypic plasticity in the persistence of extant morphotypes and the role of heritable genetic differences in establishing and maintaining these distinct cisco forms is vital for developing informed restoration strategies. To address this need, we employed a RAD sequencing approach to subsample the genomes of more than 1,300 cisco throughout the Great Lakes basin, including morphotypes C. artedi, C. hoyi, C. kiyi, C. nigripinnis, and C. zenithicus. We examined neutral and adaptive genetic differentiation, genetic diversity, and effective population sizes to identify potential conservation units and broodstock sources, and we used a newly-developed C. artedi linkage map to investigate adaptive differentiation within and among morphotypes and lakes with the goal of understanding the evolutionary mechanisms behind the phenotypic variation observed between forms.
3:40PM Genomic Tools to Define Management Units and Understand Local Adaptation in Arctic Char (Salvelinus alpinus) in Nunavik (Québec)
  Xavier Dallaire
The Arctic Char (Salvelinus alpinus) is an anadromous salmonid of great importance for traditional fisheries in the Inuit communities of Northern Canada. Arctic Char are known to migrate to coastal feeding areas in proximity to their natal rivers, which could indicate a potential for local adaptation to marine as well as freshwater conditions. In an effort to develop biologically significant management units for Arctic Char in the southwestern Canadian Arctic, we assessed genetic population structure and adaptation at a local and regional scale: 800 individuals were sampled from 25 localities in Nunavik (Quebec), Baffin Island (Nunavut) and northern Labrador, and over 25 000 SNP markers were obtained using genotyping-by-sequencing. Patterns of genetic diversity show signs of genetic structure, with pairwise Fst values between rivers ranging from 0.07 to 0.16. At similar geographic distances, populations on a same coast do not appear to be less divergent than those separated by open ocean. Association between genetic markers and environmental factors were tested using both monogenic (e.g. Bayenv, LFMM) and polygenic (e.g. RDA) methods, to identify potential targets for local adaption. These results will inform local and regional decision makers about conservation of Arctic Char stocks in Nunavik.
4:00PM Massive Data from Small Populations of the Giant Kangaroo Rat
  Mark Statham
The federally listed giant kangaroo rat (Dipodomys ingens) has been extirpated from most of its historical range due to habit loss, resulting in substantial population fragmentation. To aid in the conservation of the species we examined the evolutionary relationship among relict populations, and investigated the presence of divergence selection in different parts of the range. Using genotyping by sequencing with stringent filtering criteria we retained 35 thousand SNPs across >9 million nucleotide bases in 101 giant kangaroo rats. We identified at least five populations within the species, with varying degrees of genetic differentiation. Two of these populations were geographically isolated as well as genetically depauperate, thus suggesting that drift may have played a role in their divergence. We also identified several outlier loci among populations, most notable of which was in a homeobox gene that may be under divergent selection in southern and northern giant kangaroo rat populations. This gene regulates embryonic development and may go some way towards explaining the size differences observed within the species. Knowledge of the discrete populations within the species as well as adaptive differences will aid managers make conservation decisions especially in light of changing climate.
4:20PM Stock Structure and Contribution of West and East Basin Walleye to Recreational and Commercial Fisheries in Lake Erie
  Peter T. Euclide
Genomic techniques have been extensively used to inform mixed-stock fisheries management along North America’s west coast, however, these same techniques are only now being used to support management in the Laurentian Great Lakes basin. Throughout the basin, Walleye (Sander vitreus) is an ecologically and economically important species and the focus of extensive stock discrimination research, especially in Lake Erie. While Lake Erie’s Walleye population is supported by multiple local spawning stocks, managers have been unable to consistently discriminate among them and hence determine each stock’s relative contributions to fisheries. We present the first attempt at stock reassignment in the Great Lakes using a RAD-Capture (Rapture) panel of 12,081 polymorphic loci. We used the R package assignPOP to successfully reassign 384 individuals of known spawning origin to the lake-basin level (east vs. west). Following panel optimization, we tested the panel on 1,000 a mixed assemblage of individuals of unknown spawning-site origin to quantify the contributions of east vs. west basin stocks to eastern Lake Erie’s recreational and commercial fisheries. Our study currently is the largest genetic mixed-stock analysis of Walleye in the Great Lakes and one of the first examples of next-generation sequencing technology being used for management in the region.
4:40PM Conservation and Management in the Age of Genomics: A Cautionary Tale
  Devon Pearse
As evidenced by many of the talks in this symposium, research on the evolutionary genomic characteristics of natural populations has made spectacular progress in the past few years, largely due to the advances in sequencing technology and analysis. Without question, the resulting data have improved our ability to address numerous important evolutionary genetic questions. What is far less clear, but has been the focus of active discussion, is how adaptive genomic variation can or should inform wildlife conservation practice. Before this occurs, consideration must be given to the implications of using specific adaptive genomic targets to set conservation priorities and to the biological realities of nature populations. In this talk I will present examples associated with specific phenotypes in salmonids and other fishes to highlight the key issues and limitations for the incorporation of adaptive genomic variation in conservation and management. These examples illustrate the practical considerations and potential pitfalls of such efforts, and the importance of validating inferences drawn from new genomic data before applying them in conservation practice.

Organizers: Marlis Douglas, Michael E. Douglas, Kristin Brzeski, Emily Latch
Supported by: AFS Genetics Section, TWS Molecular Ecology Working Group

Location: Reno-Sparks CC Date: October 1, 2019 Time: 8:00 am - 5:00 pm