EST sequences were obtained from the EST database at the NCBI (http://www.ncbi.nlm.nih.gov/dbest/). A total of 76,566 ESTs were obtained for this study
(10 May 2008). All EST sequences were processed by trimming off the polyA/T and vector sequences, clustered and assembled using the TGICL software (http:// www.tigr.org/tdb/tgi/software/) to remove redundant sequences. The unique sequences were used for searches to find SSR sequences using the Troll software (http://wsmartins.net/webtroll/troll.html, now http:// wsmartins.net/websat/; Martins et al. 2009). The software was set to detect repeat motifs with more than four repeats (a mixture of type 1 and type 2 microsatellites; (Shultz et al. 2007). Both simple and complex microsatellite
motifs under 8 bp were detected. Insertion or deletions of greater than 8 bp were also detected. EST primers were designed from conserved sequences flanking repeats where the main parameters were: GC content of 40–60; annealing temperature (Tm) of 53–57C; and expected amplified products size of 100–300 bp. In some cases, candidate EST markers were prioritized based on presence of motif sizes or known polymorphisms in the EST database.