Fig. 3

a Total contamination rates for each pooling scenario. Three replicates are presented with different types of bars. Wider bars with dashed borders represent the average of the three replicates, the exact values of which are labeled on top. The exact values are shown in Additional file 1: Table S3. b index split rates when pooling was performed after PCR amplification. Average ± standard deviation (SD) of three replicates is presented. The theoretical split rate for each index is 0.125. c index contamination matrix when pooling occurred after PCR purification. Indexes 1 to 8 were assigned to Notch1, EFEMP2, Lox, USP9Y, HIST1H1D, C7orf61, GXYLT2, and TM9SF4 respectively. Read numbers and percentages are shown with or without Q30 filter application. Green shading, proper combinations; brown and yellow shading, improper combinations; yellow shading, improper combinations likely resulting from contamination during oligo synthesis. Index contamination rates were calculated by dividing the sum of contaminated reads by the sum of total reads for all eight indexes