Background <p>The Swedish Phenylketonuria screening biobank contains neonatal dried blood spots from over five million individuals born in Sweden since 1975. While its value has been proven in several research areas, the feasibility of DNA methylation profiling from decades-old samples with ultra-low DNA inputs using the Illumina Infinium MethylationEPIC v2.0 remains uncertain.</p> Methods <p>We selected samples from seven random individuals born between 1985 and 2003. We used the EPIC v2.0 for DNA methylation analysis. To asses its performance, we conducted a quality control; examining probe call rate, signal intensity, beta value density, sample-to-sample correlation, and nine control probes built into the array.</p> Results <p>Despite using DNA input quantities as low as 19.2 ng (less than 10% of manufacturer recommendations), all samples achieved high probe call rates (mean 99.75% at <i>p</i> &lt; 0.01) with robust coverage (930,588–936,110 probes), balanced methylated/unmethylated signal intensities, characteristic bimodal beta distributions, high inter-sample correlations (r = 0.972-1.000), and appropriate control probe performance. No outliers or batch effects were detected despite 18–40 years storage.</p> Conclusions <p>The EPIC v2.0 array generates high quality methylation data from archival neonatal dried blood spots with extremely low volumes of DNA input, validating the technical feasibility of large-scale epigenetic studies using the Swedish PKU biobank.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Performance of the Illumina Infinium MethylationEPIC v2.0 array with low DNA input from Swedish neonatal dried blood spots

  • Gustaf Brander,
  • Håkan Karlsson,
  • Christina Dalman,
  • Jonas Bybjerg-Grauholm,
  • James J. Crowley,
  • David Mataix-Cols

摘要

Background

The Swedish Phenylketonuria screening biobank contains neonatal dried blood spots from over five million individuals born in Sweden since 1975. While its value has been proven in several research areas, the feasibility of DNA methylation profiling from decades-old samples with ultra-low DNA inputs using the Illumina Infinium MethylationEPIC v2.0 remains uncertain.

Methods

We selected samples from seven random individuals born between 1985 and 2003. We used the EPIC v2.0 for DNA methylation analysis. To asses its performance, we conducted a quality control; examining probe call rate, signal intensity, beta value density, sample-to-sample correlation, and nine control probes built into the array.

Results

Despite using DNA input quantities as low as 19.2 ng (less than 10% of manufacturer recommendations), all samples achieved high probe call rates (mean 99.75% at p < 0.01) with robust coverage (930,588–936,110 probes), balanced methylated/unmethylated signal intensities, characteristic bimodal beta distributions, high inter-sample correlations (r = 0.972-1.000), and appropriate control probe performance. No outliers or batch effects were detected despite 18–40 years storage.

Conclusions

The EPIC v2.0 array generates high quality methylation data from archival neonatal dried blood spots with extremely low volumes of DNA input, validating the technical feasibility of large-scale epigenetic studies using the Swedish PKU biobank.