Analysis

The 84 samples from the 7 randomly assigned pairs were given genotype calls for the 96 SNPs included in the Fluidigm SNPtrace panel and for the 47 SNPs included in the iPLEX Sample ID panel. The SNPs included in the Fluidigm SNPtrace and Agena iPLEX Sample ID Plus panels are listed in Supplementary Table ². The genotype call of the pure samples for each pair was used as reference (Supplementary Table ¹). Specifically, for the mixes in which sample 1 was “contaminated” with sample 2 at different ratios, we used sample 1 as reference. Vice versa, for the mixes in which sample 2 was “contaminated” with sample 1 at different ratios, we used sample 2 as reference (Supplementary Fig. ¹).

A matrix including the genotype calls of the 84 samples assessed was generated for both panels. Concordance was defined based on the genotype similarity between the reference sample and the test sample. Specifically, the concordance between a reference sample and a test sample was calculated as the percentage of SNP assays with identical genotypes, where neither of the SNP assays gave a discrepant call; i.e. concordance = ((number of matching genotypes between test sample and reference sample)/(total number of SNPs)) * 100%. If both the reference and the test samples returned a no-call for a given SNP, that call was excluded from the calculation of concordance. The total number of SNP assays was 96 for the Fluidigm SNPtrace panel and 47 for the Agena iPLEX Sample ID Plus panel. The concordance values ranged between 0% (if no common genotypes were seen at any locus) and 100% (if the exact same genotypes were seen in both the samples at all loci). It should be noted that differently to other reports using the Tanabe algorithm to compute pairwise similarity^20,30, we used genotype similarity to calculate concordance values (Supplementary Fig. ¹)³¹. The concordance assessment in this study was based on the SNP composition, i.e. genotype match, rather than on the count of individual alleles. As an example, a homozygous A/A and a heterozygous A/B samples were considered discordant even if both samples shared the same A allele.

Pearson’s test was performed to assess the correlation between the Fluidigm SNPtrace and the Agena iPLEX Sample ID Plus concordance values. All statistical tests were based on two-tailed probability. A p-value < 0.05 was considered statistically significant. All the statistical analyses were performed using Statgraphics Centurion (V. 15, StatPoint, Inc.). Graphs were generated on Microsoft Excel for Mac, version 16.46, when relevant.