Reproducibility of the Infinium methylationEPIC BeadChip assay using low DNA amounts

Abstract

The Infinium MethylationEPIC BeadChip (EPIC) is a reliable method for measuring the DNA methylation of more than 850,000 CpG positions. In clinical and forensic settings, it is critical to be able to work with low DNA amounts without risking reduced reproducibility. We evaluated the EPIC for a range of DNA amounts using two-fold serial dilutions investigated on two different days. While the β-value distributions were generally unaffected by decreasing DNA amounts, the median squared Pearson’s correlation coefficient (R2) of between-days β-value comparisons decreased from 0.994 (500 ng DNA) to 0.957 (16 ng DNA). The median standard deviation of the β-values was 0.005 and up to 0.017 (median of medians: 0.014) for β-values around 0.6–0.7. With decreasing amounts of DNA from 500 ng to 16 ng, the percentage of probes with standard deviations ≤ 0.1 decreased from 99.9% to 99.4%. This study showed that high reproducibility results are obtained with DNA amounts in the range 125–500 ng DNA, while DNA amounts equal to 63 ng or below gave less reproducible results.

Keywords

• dna methylation
• illumina methylationepic array
• variability
• correlation
• reproducibility
• low dna amounts