Our Long Template Protocol reduces sample bleeding by accurate index assignments.

Comparison of sample bleeding in multiplexed samples, sequenced with either standard Illumina protocol (“Illumina”) or our Long Template Protocol (“Our”). Columns: Percentage of non-human sample (top table shows results for yeast, bottom table for PhiX), template protocol, multiplexed (yes or no), observed sample bleeding, expected sample bleeding, observed to expected ratio. Sample bleeding was estimated as a percentage of reads in the human sample that do not map to the human genome, but map without mismatches to the control organism genome (see Materials and Methods, in the last two rows of each samples that are 100% human estimated sample bleeding is the observed false positive frequency for our test). Note that sample bleeding is the highest for Illumina Standard Protocol and that Long Template Protocol can reduce sample bleeding by 2-5 times as compared to Standard Illumina Protocol, although still remains substantial.Our Long Template Protocol reduces sample bleeding by accurate index assignments.