First, we randomly divide the sequencing data into several partitions. We chose to create 6 partitions from each of the 3 libraries (18 partitions total), therefore creating data partitions with ~5x each. We accomplished this by sorting the BAM by name using SortSam from the Picard (http://picard.sourceforge.net) tools to effectively give the reads random ordering. We then randomly allocate each read to one of the partitions and write it to a partition-specific BAM file.
