Elsevier

Aquaculture

Volume 491, 1 April 2018, Pages 105-113
Aquaculture

Transcriptome based SNP discovery and validation for parentage assignment in hatchery progeny of the European abalone Haliotis tuberculata

https://doi.org/10.1016/j.aquaculture.2018.03.006Get rights and content

Highlights

  • This is the first SNP panel developed for parentage assignment in an abalone species.

  • SNP discovery used an existing transcriptome, and yielded 298 SNPs with > 90% call success across >1000 genotyped abalone

  • A subset of 123 SNPs successfully assigned parentage in 98.9% of 945 offspring (189 mixed families from 40 parents).

  • Parentage assignment revealed that broodstock reproductive success was highly variable.

  • Careful management of genetic variability would be needed to implement individual selection in hatchery produced progenies.

Abstract

Selective breeding strategies require pedigree information over generations, but many species produced in aquaculture are too small to be physically tagged at early stages. Consequently, maintaining a sufficient number of separate families is often needed but costly and logistically difficult. Alternatively, parentage assignment can be obtained using DNA markers. We developed a panel of single nucleotide polymorphism (SNP) markers for the European abalone Haliotis tuberculata using an existing transcriptomic resource. An initial set of 2,176,887 SNPs was filtered to select 500 for high throughput genotyping. Of these, 298 SNPs were amplified in at least 90% of our H. tuberculata samples, consisting of a mixed family cohort (945 offspring) generated by crossing 40 abalones, and 5 full-sib training families (70 offspring). Based on amplification success among parents, minimum allele frequency and checks carried out against the training families, a subset of 123 markers was used to carry out parentage assignment in our mixed family cohorts. Maximum likelihood and exclusion-based methods of parentage assignment yielded consistent results, allowing parentage to be assigned in 98.9% of the studied progeny. Optimization of markers suggests that the 60 most informative SNPs may be sufficient for 95% assignment success in these progeny. The panel was also used to estimate effective population size, and revealed a low Ne due to high variance of reproductive success between parents. Our panel could be used to estimate genetic parameters of traits in mixed family cohorts, an essential stage to initiate selective breeding in H. tuberculata. It could also be useful tool in the context of monitoring stock enhancement and population genetics studies.

Introduction

Abalone production is an emerging industry in Europe, and is mainly based on the European abalone Haliotis tuberculata (Huchette and Clavier, 2004), although Haliotis discus hannai and Haliotis rufescens have been introduced to Ireland (Hannon et al., 2013) and Iceland (Jonasson et al., 1999) respectively for aquaculture production. H. tuberculata takes between 3 and 4 years to reach a minimal commercial size of 30 g, and no intentional selective process has yet been initiated (Lachambre et al., 2017). This slow growth leads to high production costs and increases the risk of accidents during the production cycle. To reduce growing time, many producers and breeders have adopted selective breeding programs for the different abalone species farmed worldwide (Elliott, 2000). These programs require considerable investment in the long run, including dedicated research projects, special rearing facilities, precise traceability and phenotyping costs (Gjedrem and Baranski, 2010). Furthermore, as with many aquatic species, abalone are too small to be physically tagged in the early stage of production, which can complicate the implementation of family-based selective breeding programs (Vandeputte and Haffray, 2014). Either all the families are mixed at the larval stage, with individuals selected only on the basis of their phenotype at the end of the rearing without pedigree information, or all the families are reared separately until physical tagging. Individual selection can generate improved phenotypes, but cannot control for inbreeding unless multiple cohorts are selected and intercrossed. Rearing families separately can control inbreeding, but is costly and heritability estimates may be influenced by common environmental conditions between tanks (Gjedrem and Baranski, 2010).

Alternatively, DNA markers can be used to identify the pedigree of animals reared in mixed families under commercial conditions; parentage assignment with DNA markers is thus an invaluable tool for assessing the extent of inbreeding and can help to optimise the breeding strategy during the domestication process (Vandeputte and Haffray, 2014). Until recently, microsatellites were the most commonly used genetic marker for parentage analysis in non-model organisms (Jones et al., 2010; Weinman et al., 2015), including species of interest for aquaculture (Liu et al., 2017; Trọng et al., 2013). Among different abalone species, several studies have used microsatellites to infer pedigree, resulting in up to 90–95% assignment success (Hara and Sekino, 2007; Lucas et al., 2006; van den Bergb and Roodt-Wilding, 2010). Yet recent advances in gene sequencing and genotyping technologies have allowed the use of alternative markers such as single nucleotide polymorphisms (SNPs) in parentage assignment. Such markers display a number of advantages over traditional microsatellite markers, including better assignment (Sellars et al., 2014; Trọng et al., 2013), more accurate genotyping (Anderson and Garza, 2006), fewer null alleles (van den Bergb and Roodt-Wilding, 2010) and the potential for quantitative trait loci (QTL) detection (Avia et al., 2017; Gutierrez et al., 2014). Furthermore, SNPs can be derived from increasingly common expressed sequence data such as the short reads and transcriptomes generated during RNA-Seq experiments (De Wit et al., 2015), making the development of novel markers straightforward compared to microsatellites. Non-synonymous substitutions in SNPs from coding regions may help to identify candidate genes for marker assisted selection (Merwe et al., 2013), and more broadly can be used in genetic analyses to understand the structure of the population and manage the stock (Rhode et al., 2017).

The number of markers required for successful parentage assignment can be quite variable, depending on the heterozygosity of the markers in progenies. Liu et al. (2017) found that 50 SNPs allowed 100% assignment in the Pacific oyster Crassostrea gigas; yet in another study of a different population of C. gigas, Lapègue et al. (2014) found that 150 SNPs were necessary for unambiguous parentage assignment. While assignment success for SNP panels is clearly influenced by numerous population genetic factors and by the quality of the markers, optimization approaches based on using the most polymorphic markers (Perez-Enriquez and Max-Aguilar, 2016) or testing subsets of SNPs from a larger panel (Holman et al., 2017) can help to determine smaller optimal panels for particular populations. The assignment success of a panel may also be influenced by the method used to achieve it. Numerous methodologies for assigning parentage exist (Jones et al., 2010), but among the most widely used are exclusion-based methods and maximum likelihood-based methods. Exclusion-based methods are sensitive to genotyping errors whereas likelihood-based methods generally give higher assignment rates (especially with low power marker sets), but can give inconsistent results (Vandeputte and Haffray, 2014). Direct comparisons of the two methods (e.g. Perez-Enriquez and Max-Aguilar, 2016; Trọng et al., 2013) showed similar levels of assignment were achieved with the same number of markers. Furthermore, Trọng et al. (2013) found that likelihood-based methods could help resolve certain cases in which exclusion methods assigned multiple parents. Comparing the results from multiple methods can increase assignment confidence (Morvezen et al., 2013), and help to determine the optimal number of markers necessary to assign parentage in a given population.

In the context of selective breeding programs, an important output of an SNP panel is its capacity to estimate effective size (Ne) of populations under selection (van den Bergb and Roodt-Wilding, 2010), which is an important indicator of inbreeding. The Ne of cultured populations is often lower than the number of parents, a fact that is especially true for aquatic species which are characterized by high fecundity and high variability in reproductive success (Boudry et al., 2002; Hedgecock, 1994). In closed populations such as selected farm populations, low Ne (i.e. high genetic drift) can lead to a deleterious loss of diversity (Falconer and Mackay, 1996). If no temporal information is available, Ne can be calculated from the variability of reproduction success of the different parents contributing to a cohort (Chevassus, 1989). Ne can also be estimated from genetic parameters such as heterozygotes excess (Zhdanova and Pudovkin, 2008) or linkage disequilibrium (Waples and Do, 2008). Methods for estimating Ne with molecular markers are still being developed and tested (Waples, 2016), and different methods applied to the same data set often result in different estimates of Ne (Morvezen et al., 2016; Wang, 2016; Waples and Do, 2010); however, comparison of these values can provide a measure of confidence, particularly if they are based on independent information (Waples, 2016). In abalone hatcheries, Ne has been estimated at 18 to 75 for H. rubra farms in Australia (Evans et al., 2004), and, at 15–67 for H. midae following two generations of selection on farms in South Africa (Rhode et al., 2014). In H. midae, these Ne estimates were associated with elevated rates of inbreeding (5%), and assessing Ne is thus an important step in understanding loss of genetic diversity when implementing new breeding practices.

The first aim of this study was to design a panel of SNPs using data from a transcriptomic database for high throughput parentage assignment. The second aim was to test the panel's utility for parentage assignment in a real cohort reared during three years at an abalone farm. Finally the inferred pedigree was used to estimate Ne in this cohort.

Section snippets

Rearing of mixed families cohorts and bi-parental families

A total of 945 abalone were reared during 32 months at France Haliotis (48°36′46 N, 4°33′30 W; Plouguerneau, France), a commercial French hatchery. The studied progeny resulted from a full factorial mating of 24 males and 16 females randomly sampled from France Haliotis farmed stock. Two epipodia (approximately 50 mg of tissue) were sampled from each parent after the spawning and stored in 70% ethanol. Farm stock from France Haliotis was not under intentional selective pressure and animals were

SNP discovery and filtration in silico

The variant call file (vcf) output by samtools identified a total of 2,176,887 SNPs in the H. tuberculata transcriptome. The initial heavy filtration removed nearly 99% of these, leaving 2368 SNPs across 1923 contigs, all of which had phred scaled quality scores above 200 and read depths between 150 and 310. Transdecoder identified 556 contigs with at least one ORF (a total of 986 ORFs), which in combination with blastp results allowed the sense of the contig to be resolved in most cases. Just

Discussion

In this study we successfully used a previously published transcriptomic dataset to generate a panel of informative SNPs for parentage assignment of the European abalone H. tuberculata. Comparing maximum-likelihood and exclusion-based methods to assign parentage yielded similar results and showed that a panel consisting of as few as 60 SNPs was enough to achieve >95% assignation success. Parentage assignment revealed high variance in reproductive success, which when taken with the Ne estimate

Conclusion

Here we present a novel panel of 298 SNP markers for the European abalone H. tuberculata. The panel was developed using an existing transcriptomic database and demonstrated comparable success rates to SNP discovery experiments based on intentionally collected genomic data. A subset of 123 markers had a theoretical potential assignment of 100%, and provided 98.9% assignment in a real farm cohort. The 60 most-informative markers were sufficient to achieve more that 95% assignment. Parentage

Acknowledgements

The authors would like to thank Xavier Lesage, Frederic Laurans, Michael Gleeson, Iain McKensy, and Maryvonne Leroux, from the production team of France Haliotis, for the 4-year rearing of the mixed cohort families. Abims (Analysis Bioinformatics for Marine Science, Roscoff Marine Station) provided server space for bioinformatic analyses. In addition, the authors would like to thank Anastasia Bestin and Anne-Sophie Tiriau from the SYSAAF for help with phenotyping, and Thomas Bisch and Amy

References (73)

  • S. Lachambre et al.

    Relationships between growth, survival, physiology and behaviour — a multi-criteria approach to Haliotis tuberculata phenotypic traits

    Aquaculture

    (2017)
  • T. Lucas et al.

    Heritability estimates for growth in the tropical abalone Haliotis asinina using microsatellites to assign parentage

    Aquaculture

    (2006)
  • T.T.T. Nguyen et al.

    Genetic parameters and response to selection in blue mussel (Mytilus galloprovincialis) using a SNP-based pedigree

    Aquaculture

    (2014)
  • R. Nielsen et al.

    Correcting for ascertainment biases when analyzing SNP data: applications to the estimation of linkage disequilibrium

    Theor. Popul. Biol.

    (2003)
  • H.R. Nolasco-Alzaga et al.

    Quantitative genetic parameters of growth and fatty acid content in the hemolymph of the Whiteleg shrimp Litopenaeus vannamei

    Aquaculture

    (2018)
  • T.Q. Trọng et al.

    A comparison of microsatellites and SNPs in parental assignment in the GIFT strain of Nile tilapia (Oreochromis niloticus): the power of exclusion

    Aquaculture

    (2013)
  • E.C. Anderson et al.

    The power of single-nucleotide polymorphisms for large-scale parentage inference

    Genetics

    (2006)
  • K. Avia et al.

    High-density genetic map and identification of QTLs for responses to temperature and salinity stresses in the model brown alga Ectocarpus

    Sci. Rep.

    (2017)
  • A. Bilbao et al.

    Effect of macroalgae protein levels on some reproductive aspects and physiological parameters for the abalone, Haliotis tuberculata coccinea (Reeve 1846)

    J. World Aquacult. Soc.

    (2012)
  • M. Boulais et al.

    Assessment of oocyte and trochophore quality in Pacific oyster, Crassostrea gigas

    Aquaculture

    (2015)
  • B. Chevassus

    Aspects génétiques de la constitution de populations d'élevage destinées au repeuplement

    Bull. Français la Pêche la Piscic.

    (1989)
  • S. Daume et al.

    Settlement, early growth and survival of Haliotis rubra in response to different algal species

    J. Appl. Phycol.

    (2000)
  • P. De Wit et al.

    SNP genotyping and population genomics from expressed sequences — current advances and future possibilities

    Mol. Ecol.

    (2015)
  • A.B. der Merwe et al.

    Transcriptome-wide single nucleotide polymorphisms (SNPs) for abalone (Haliotis midae): validation and application using GoldenGate medium-throughput genotyping assays

    Int. J. Mol. Sci.

    (2013)
  • C. Do et al.

    NeEstimator v2: re-implementation of software for the estimation of contemporary effective population size (Ne) from genetic data

    Mol. Ecol. Resour.

    (2014)
  • N.G. Elliott

    Genetic improvement programmes in abalone: what is the future?

    Aquac. Res.

    (2000)
  • D.S. Falconer et al.

    Introduction to Quantitative Genetics

    (1996)
  • T. Gjedrem et al.

    Selective Breeding in Aquaculture: An Introduction

    (2010)
  • A. Gonçalves da Silva et al.

    SNP discovery in nonmodel organisms: strand bias and base-substitution errors reduce conversion rates

    Mol. Ecol. Resour.

    (2015)
  • A.P. Gutierrez et al.

    Detection of quantitative trait loci (QTL) related to grilsing and late sexual maturation in Atlantic salmon (Salmo salar)

    Mar. Biotechnol.

    (2014)
  • B.J. Haas et al.

    De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis

    Nat. Protoc.

    (2013)
  • C. Hannon et al.

    Review of the technical challenges facing aquaculture of the European abalone Haliotis tuberculata in Ireland

    Aquac. Int.

    (2013)
  • M. Hara et al.

    Parentage testing for hatchery-produced abalone Haliotis discus hannai based on microsatellite markers: preliminary evaluation of early growth of selected strains in mixed family farming

    Fish. Sci.

    (2007)
  • E.D. Harney et al.

    SNP Discovery and Validation for Parentage Assignment in Hatchery Progeny of the European Abalone Haliotis tuberculata

    (2017)
  • D. Hedgecock

    Does variance in reproductive success limit effective population sizes of marine organisms

  • D. Hedgecock et al.

    Sweepstakes reproductive success in highly fecund marine fish and shellfish: a review and commentary

    Bull. Mar. Sci.

    (2011)
  • Cited by (15)

    • No evident effect of domestication on the anti-predator behaviour of European abalone (Haliotis tuberculata): Implications for stock enhancement programs

      2021, Applied Animal Behaviour Science
      Citation Excerpt :

      Although the domestication of most aquatic organisms has only been actively pursued for 50 years, these domestication efforts quickly rendered results in some species due to a great potential for selection (Lorenzen et al., 2012). Domestication can reduce the genetic variability of organisms, as offspring are often obtained from a few spawning adults, resulting in genetic drift (Price, 1984; Grant et al., 2017; Harney et al., 2018). Furthermore, farm conditions and productivity targets (e.g. selection of fast-growing individuals) both result in intentional or unintentional selection.

    • Development and validation of a SNP-based genotyping tool for pedigree establishment in Australian greenlip abalone Haliotis laevigata Donovan, 1808

      2021, Aquaculture Reports
      Citation Excerpt :

      Aquaculture production of abalone is challenged by slow growth rates, supply chain mortality and high production costs (Reaburn and Edwards, 2003); accordingly, there is much interest within the industry to develop and implement breeding programs to achieve sustainable productivity gain and improve farm return via genetic improvement (AAGA, 2020). Recently, several genetic marker panels have been developed for various abalone species illustrating their applications in determining natural population structure and for pedigree construction in breeding programs (Harney et al., 2018; Kijas et al., 2019; Miller et al., 2019; Sandoval-Castillo et al., 2018). Effective design and optimal performance of SNPs included on a parentage panel relies on validating the SNPs within the genetic diversity (i.e. farm broodstock) they are intended to be genotyped to reduce ascertainment biases.

    View all citing articles on Scopus
    View full text