Detection of molecular signatures of selection at microsatellite loci in the South African abalone (Haliotis midae) using a population genomic approach Clint Rhode ⁎, Jessica Vervalle, Aletta E. Bester-van der Merwe, Rouvay Roodt-Wilding Department of Genetics, Stellenbosch University, Private Bag X1, Matieland, 7602, South Africa abstract article info Article history: Received 12 July 2012 Received in revised form 12 March 2013 Accepted 17 March 2013 Keywords: Adaptation F st -outlier Linkage disequilibrium Neutrality Population genomics Selection Identifying genomic regions that may be under selection is important for elucidating the genetic architecture of complex phenotypes underlying adaptation to heterogeneous environments. A population genomic approach, using a classical neutrality test and various F st -outlier detection methods was employed to evaluate genome-wide polymorphism data in order to identify loci that may be candidates for selection amongst six populations (three cultured and three wild) of the South African abalone, Haliotis midae. Approximately 9% of the genome-wide microsatellite markers were putatively subject to directional selection, whilst 6–18% of the genome is thought to be influenced by balancing selection. Genetic diversity estimates for candidate loci under directional selection was significantly reduced in comparison to candidate neutral loci, whilst can- didate balancing selection loci demonstrated significantly higher levels of genetic diversity (Kruskal–Wallis test, P b 0.05). Pairwise F st estimates based on candidate directional selection loci also demonstrated in- creased levels of differentiation between study populations. Various candidate loci under selection showed significant inter-chromosomal linkage disequilibrium, suggesting possible gene-networks underling adaptive phenotypes. Furthermore, several loci had significant hits to known genes when performing BLAST searches to NCBI's non-redundant databases, whilst others are known to be derived from expressed sequences even though homology to a known gene could not be established. A number of loci also demonstrated relatively high similarity to transposable elements. The association of these loci to functional and genomically active sequences could in part explain the observed signatures of selection. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Understanding how selection shapes molecular diversity and how this diversity in turn facilitates the development of phenotypes, in heterogeneous environments, has become a key endeavour of modern evolutionary biology. Recently, population genomic scans have become increasingly popular for detecting population divergence as a conse- quence of adaptation and identifying the underlying genetic architecture of complex divergent phenotypes (Pavlidis et al., 2008; Stinchcombe and Hoekstra, 2008; Nielsen et al., 2009a; Nosil et al., 2009). The use of genome-wide polymorphism data allows for the partitioning of locus-specific effects such as recombination, mutation and selection from demographic effects (including: bottlenecks, founder effects, popu- lation stratification and migration etc.) (Luikart et al., 2003; Stinchcombe and Hoekstra, 2008) and also provides for functional analyses of genetic polymorphisms to be extrapolated to a population level (Bonin, 2008). Unlike the more conventional linkage-based QTL analysis, population genomic scans do not rely on structured pedigree information and controlled breeding experiments, often impractical when working with natural populations or organisms with long generation times (Storz, 2005; Stinchcombe and Hoekstra, 2008). Furthermore, there is no dependence on a priori phenotypic information as is the case for association- and conventional linkage-based studies. This is particularly advantageous when phenotypic traits are not readily observable or unknown, such as biochemical or physiological traits (Storz, 2005; Walsh, 2008). Recently, F st -outlier tests became popular, because it allowed for the simultaneous analysis of a large number of loci and both domi- nant and co-dominant marker data could be employed. F st -outlier tests have been used in a number of studies in order to detect loci that might be under selection (e.g. Bonin et al., 2006; Paris et al., 2010; Nunes et al., 2011) including those for aquatic species, such as, Atlantic salmon (Salmo salar; Vasemägi et al., 2005), lake whitefish (Coregonus clupeaformis; Campbell and Bernatchez, 2004), cod (Gadus morhua; Nielsen et al., 2009b), Eastern oyster (Crassostrea virginica; Murray and Hare, 2006) and periwinkle snail (Littorina saxatilis; Wilding et al., 2001) (for a review: Nielsen et al., 2009a). The South African abalone, locally known as “perlemoen” (Haliotis midae Linnaeus, Haliotidae; Gastropoda), is an economically impor- tant marine mollusc. The species has a wide distribution along the cool to warm temperate regions of the South African coast. Although historically an important fisheries species, the commercial sector currently relies mainly on aquaculture due to the suspension of Marine Genomics 10 (2013) 27–36 ⁎ Corresponding author. Tel.: +27 21 808 5837; fax: +27 21 808 5833. E-mail address: clintr@sun.ac.za (C. Rhode). 1874-7787/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.margen.2013.03.001 Contents lists available at SciVerse ScienceDirect Marine Genomics journal homepage: www.elsevier.com/locate/margen