7 Genomic Approaches in Aquaculture and Fisheries
217
an increasing number of available EST-SSR (e.g. Yu and Li 2008). Another possibility to position genes on a genetic linkage map is to use SNP loci. While SSR loci are
well established genetic markers, SNPs represent a recent addition to the genomic
toolbox for most animal species. So far, the representation of SNP markers in linkage maps of marine species has been limited to few loci but the development of high
throughput SNP genotyping should allow a significant increase in the number of
mapped SNPs in the near future. Compared to SSR loci, SNPs are more frequent in
the genome, both in coding and non-coding regions, and are bi-allelic, which makes
them portable across laboratories, and more amenable to high-throughput analysis and automation. The frequency of SNPs is particularly high in marine bivalves
(e.g. one SNP every 60 bp in coding regions and one every 40 bp in non-coding
regions; Sauvage et al. 2007), which have high effective populations sizes (i.e. the
number of effective breeders at each generation). SNPs are indeed less polymorphic
than SSRs, but their frequency in most species and the implementation of highthroughput technologies makes possible the identification of hundreds of thousands
SNPs. Likewise, the genotyping of up to 1,000,000 SNP loci is now possible (e.g.
Illumina Infinium HD Human 1 M). At a much smaller scale, SNP isolation and
genotyping is well advanced, especially in the Atlantic salmon (Hayes et al. 2007),
while other marine species are quickly progressing (e.g. Atlantic cod, European sea
bass and Pacific oyster).
Whatever the type of marker used for the construction of a linkage map, it is necessary to sample a large number of independent meiotic events in order to measure
recombination between loci. Traditionally, this has been achieved through the use of
dedicated mapping panels, i.e. experimental populations originating from F1, backcross or F2 crosses between highly divergent strains. As such populations are not yet
available for most marine fish and shellfish, mapping panels are often the F1 generation from a cross between two wild animals. Despite these limitations, the high
levels of heterozygosity observed in most marine species (having very large effective population sizes) increases the probability of each marker being informative for
the construction of the map.
Linkage maps represent a scaffold of markers for genomic studies. These markers can be used in genome-wide scans for linkage with phenotypic characters of
economic or scientific interest. Apart from the identification of Quantitative Trait
Loci (QTL) in farmed species, one excellent example of the potential of such
methodology comes from the identification of markers linked to loci responsible
for morphological divergence between sympatric populations of three-spined stickleback. Peichel et al. (2001) used a medium resolution linkage map to analyze the
genetic basis of recently evolved changes in skeletal armour and feeding morphologies seen in the benthic and limnetic stickleback species from Priest Lake, British
Columbia.
The main limitations of currently available linkage maps are the scarce representation of coding genes among mapped loci and the low level of comparability
between linkage maps from different species. Low-to-moderate resolution linkage
maps have been developed during the last decade for several fish and shellfish of
aquacultural importance (for review, see Wenne et al. 2007). Most of the maps currently available are based on F1 crosses and AFLP markers. Many of them are first
217
an increasing number of available EST-SSR (e.g. Yu and Li 2008). Another possibility to position genes on a genetic linkage map is to use SNP loci. While SSR loci are
well established genetic markers, SNPs represent a recent addition to the genomic
toolbox for most animal species. So far, the representation of SNP markers in linkage maps of marine species has been limited to few loci but the development of high
throughput SNP genotyping should allow a significant increase in the number of
mapped SNPs in the near future. Compared to SSR loci, SNPs are more frequent in
the genome, both in coding and non-coding regions, and are bi-allelic, which makes
them portable across laboratories, and more amenable to high-throughput analysis and automation. The frequency of SNPs is particularly high in marine bivalves
(e.g. one SNP every 60 bp in coding regions and one every 40 bp in non-coding
regions; Sauvage et al. 2007), which have high effective populations sizes (i.e. the
number of effective breeders at each generation). SNPs are indeed less polymorphic
than SSRs, but their frequency in most species and the implementation of highthroughput technologies makes possible the identification of hundreds of thousands
SNPs. Likewise, the genotyping of up to 1,000,000 SNP loci is now possible (e.g.
Illumina Infinium HD Human 1 M). At a much smaller scale, SNP isolation and
genotyping is well advanced, especially in the Atlantic salmon (Hayes et al. 2007),
while other marine species are quickly progressing (e.g. Atlantic cod, European sea
bass and Pacific oyster).
Whatever the type of marker used for the construction of a linkage map, it is necessary to sample a large number of independent meiotic events in order to measure
recombination between loci. Traditionally, this has been achieved through the use of
dedicated mapping panels, i.e. experimental populations originating from F1, backcross or F2 crosses between highly divergent strains. As such populations are not yet
available for most marine fish and shellfish, mapping panels are often the F1 generation from a cross between two wild animals. Despite these limitations, the high
levels of heterozygosity observed in most marine species (having very large effective population sizes) increases the probability of each marker being informative for
the construction of the map.
Linkage maps represent a scaffold of markers for genomic studies. These markers can be used in genome-wide scans for linkage with phenotypic characters of
economic or scientific interest. Apart from the identification of Quantitative Trait
Loci (QTL) in farmed species, one excellent example of the potential of such
methodology comes from the identification of markers linked to loci responsible
for morphological divergence between sympatric populations of three-spined stickleback. Peichel et al. (2001) used a medium resolution linkage map to analyze the
genetic basis of recently evolved changes in skeletal armour and feeding morphologies seen in the benthic and limnetic stickleback species from Priest Lake, British
Columbia.
The main limitations of currently available linkage maps are the scarce representation of coding genes among mapped loci and the low level of comparability
between linkage maps from different species. Low-to-moderate resolution linkage
maps have been developed during the last decade for several fish and shellfish of
aquacultural importance (for review, see Wenne et al. 2007). Most of the maps currently available are based on F1 crosses and AFLP markers. Many of them are first