As attrition of telomeres, DNA caps that protect chromosome integrity, is accelerated by various forms of stress, telomere length (TL) has been proposed as an indicator of lifetime accumulated stress. In ecological studies, it has been used to provide insights into ageing, life history trade-offs, the costs of reproduction and disease. qPCR is a high-throughput and cost-effective tool to measure relative TL (rTL) that can be applied to newly collected and archived ecological samples. However, qPCR is susceptible to error both from the method itself and pre-analytical steps. Here, repeatability was assessed overall and separately across multiple levels (intra-assay, inter-assay and inter-extraction) to elucidate the causes of measurement error, as a step towards improving precision. We also tested how accuracy, defined as the correlation between the "gold standard" for TL estimation (telomere restriction fragment length analysis with in-gel hybridization), could be improved. We find qPCR repeatability (intra- and inter-assay levels) to be at similar levels across three common storage media (ethanol, Longmire's and Queen's). However, inter-extraction repeatability was 50% lower for samples stored in Queen's lysis buffer, indicating storage medium can influence precision. Precision as well as accuracy could be increased by estimating rTL from multiple qPCR reactions and from multiple extractions. Repetition increased statistical power equivalent to a 25% (single extraction analysed twice) and 17% (two extractions) increase in sample size. Overall, this study identifies novel sources of variability in high-throughput telomere quantification and provides guidance on sampling strategy design and how to increase rTL precision and accuracy.
Keywords: molecular aging; quantitative PCR; repeatability; statistical power; telomeres.
© 2017 John Wiley & Sons Ltd.