Summary of the metagenomic data for the S. bongori ten-fold dilution series (initial undiluted samples contained approximately 108cells), extracted with four different kits. Each column represents a single sample. A sample of ultrapure water, without DNA extraction, was also sequenced (labelled `water'). a) As the starting material becomes more diluted, the proportion of sequenced reads mapping to the S. bongori reference genome decreases for all kits and contamination becomes more prominent. b) The profile of the non-Salmonella reads (grouped by Family, only those comprising >1% of reads from at least one kit are shown) is different for each of the four kits.