SSR‐seq: Genotyping of microsatellites using next‐generation sequencing reveals higher level of polymorphism as compared to traditional fragment size scoring

Abstract

Microsatellites (or simple sequence repeats, SSR) are widely used markers in population genetics. Traditionally, genotyping was and still is carried out through recording fragment length. Now, next‐generation sequencing (NGS) makes it easy to obtain also sequence information for the loci of interest. This avoids misinterpretations that otherwise could arise due to size homoplasy. Here, an NGS strategy is described that allows to genotype hundreds of individuals at many custom‐designed SSR loci simultaneously, combining multiplex PCR, barcoding, and Illumina sequencing. We created three different datasets for which alleles were coded according to (a) length of the repetitive region, (b) total fragment length, and (c) sequence identity, in order to evaluate the eventual benefits from having sequence data at hand, not only fragment length data. For each dataset, genetic diversity statistics, as well as F ST and R ST values, were calculated. The number of alleles per locus, as well as observed and expected heterozygosity, was highest in the sequence identity dataset, because of single‐nucleotide polymorphisms and insertions/deletions in the flanking regions of the SSR motif. Size homoplasy was found to be very common, amounting to 44.7%–63.5% (mean over all loci) in the three study species. Thus, the information obtained by next‐generation sequencing offers a better resolution than the traditional way of SSR genotyping and allows for more accurate evolutionary interpretations.

Application of a SSR‐GBS marker system on investigation of European Hedgehog species and their hybrid zone dynamics

“…There are alternative approaches based on the assembly of the amplicon reads. Šarhanová, Pfanzelt, Brandt, Himmelbach, and Blattner () applied an alternative approach based on read de novo assembly. Nevertheless, a manual control step was added to account for the assembly of two alleles filtering noise.…”

“…Amplicon sequencing has a special role in SSR analysis (de Barba et al, 2017;Farrell, Carlsson, & Carlsson, 2016;Vartia et al, 2016;Šarhanová et al, 2018), and microsatellite amplification is the method of choice for population genetics, due to the ability to recover multiple alleles per locus, resulting in a high statistical power with a low number of sequenced markers (Ellegren, 2004;Schlotterer, 2000). Despite the obvious advantages of whole-genome sequencing approaches, genotyping-specific loci is more cost-effective and more easily implemented, which is also one of the arguments found in recent reviews for the use of microsatellites in place of RAD/GBS (Hodel et al, 2017(Hodel et al, , 2016.…”

“…Using this method, it is now possible to recover the complete sequence composition of the locus, including the repeat motif and SNPs in the flanking region. This approach makes it possible to overcome homoplasy characteristics of microsatellites (Vartia et al, 2016;Šarhanová et al, 2018). In these cases, shared alleles resulting from homoplasy would have the same number of repetitions but different flanking regions.…”

Tabula rasa in the Patagonian Channels II: The cushion peat bog species Astelia pumila (Asteliaceae) and Donatia fascicularis (Stylidiaceae) survived the last glacial in south‐central Chile

“…Qdd 3.1 (Meglécz et al., ) was used for SSR identification in A. pumila , and Phobos 3.3.12 (Mayer, ) in the case of D. fascicularis . SSR amplicons were sequenced on an Illumina MiSeq (for details regarding sequencing and data analyses, see Šarhanová, Pfanzelt, Brandt, Himmelbach, & Blattner, ). The individuals of A. pumila and D. fasciularis had maximally two alleles per locus and were consequently scored as effective diploids, although chromosome numbers point to tetraploidy.…”

“…Since length mutations in the repetitive regions occur at a higher rate than nucleotide changes in the flanking regions, a second dataset was created based on SNP/indel variation in the flanking regions only, in order to avoid problems related to homoplasy caused by convergent SSR evolution. However, the overall picture did not change, so the original allele coding, offering a higher resolution, was used (see Šarhanová et al., ).…”