Figure 1 .
AllelePipe workflow for identifying alleles without a reference genome. Unigenes from all individuals are pooled and clustered by similarity. Clustered sequences are aligned and consensus sequences are generated, providing a pseudo-reference genome. Unigenes from the same and/or different individuals are aligned to the reference, and SNPs are identified. Multilocus SNP information is used to construct a minimum set of haplotypes for each individual, and clusters in which individuals are represented by an excess number of putative alleles are flagged as potential multigene clusters.