Sfftools

PCRs were performed in triplicate to minimize the inherent amplification bias of any given PCR of multicopy loci. The PCR products were verified by gel electrophoresis and, based on the intensity of the products, pooled in equal amounts for each of the three reactions. Each pool had a unique multiplex identifier (MID) (Roche Life Sciences) ligated to the products, which allowed for computational sorting of reads by animal post-sequencing. A 1/8 plate 454 FLX Titanium run was used to generate sequence data. The 454 sequence reads generated in this study were separated by MID using sfftools (Roche Life Sciences) for standard MIDs. Low quality reads were excluded from the analysis, resulting in a data set of approximately 103,761 reads.

BAC clones were sequenced using Roche GS FLX (Basel, Switzerland). BAC DNAs were prepared in a 96-deepwell plate, extracted using the alkaline lysis method and purified with the MultiScreen Filter plate (Millipore, Billerica, MA, USA). After fragmentation with the Covaris Acoustic Solubilizer LE220 (Covaris, Woburn, MA, USA), the sequencing libraries were constructed using the NEBNext DNA Library Prep Master Mix Set for 454 (New England Biolabs, Ipswich, MA, USA) and tagged with Multiplex Identifier (MID) Adaptors (Roche, Basel, Switzerland). The tagged DNAs were subjected to size selection using agarose gel electrophoresis to obtain DNA fragments of the appropriate size (600–1200 bp) and then quantified using the Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA). The resulting libraries were sequenced according to the manufacturer’s recommendations. The resulting SFF files were split into separate files of each BAC clone based on MID by using the sffile program in SFF Tools (Roche, Basel, Switzerland). The split SFF files were used for de novo assembly with the GS De Novo Assembler v2.6 (Roche, Basel, Switzerland) with default parameters, and the BAC vector and E. coli genome sequences were trimmed.