Methylation reference datasets from quartet DNA materials for benchmarking epigenome sequencing

Area of Science:

• Epigenetics
• Genomics
• Molecular Biology

Background:

• Lack of quantitative methylation reference datasets and reproducibility assessment impedes clinical use of epigenome-wide sequencing.
• Standardized quality control is crucial for reliable epigenomic data.

Purpose of the Study:

• To generate quantitative methylation reference datasets using certified DNA materials.
• To assess cross-laboratory reproducibility of mainstream epigenome-sequencing protocols.
• To establish foundational standards for benchmarking epigenomic technologies.

Main Methods:

• Generated 108 epigenome-sequencing datasets using whole-genome bisulfite sequencing, enzymatic methyl-seq, and TET-assisted pyridine borane sequencing.
• Utilized certified Quartet DNA reference materials with triplicates across multiple laboratories.
• Employed consensus voting to construct genome-wide quantitative methylation reference datasets.

Main Results:

• Observed strand-specific methylation biases across all tested protocols.
• Demonstrated high agreement in quantitative methylation levels (mean PCC = 0.96) but low detection concordance (mean Jaccard index = 0.36) between laboratories.
• Identified strong correlations between technical parameters (depth, coverage, strand consistency) and quality metrics (recall, PCC, RMSE).

Conclusions:

• Established foundational standards for benchmarking epigenomic technologies and analytical pipelines.
• Enabled robust and standardized quality control for epigenomic research and clinical applications.
• Facilitated the clinical translation of epigenome-wide sequencing technologies through reliable reference datasets.