Workflow used for SNP discovery in potato transcriptomes and design of the BeadXpress SNP array. SNPs identified in RNA-Seq reads were called and filtered using the Maq SNP pipeline. Sanger ESTs were clustered by cultivar using TGICL  and SNPs called and filtered using custom Perl scripts. Filtered SNPs were linked to positions of the potato DM genomic sequence and filtered again to eliminate those close to an intron as well SNPs that were not biallelic. SNPs selected for the BeadXpress SNP array were selected randomly from the Atlantic, Premier Russet, and Snowden datasets.