Performance comparisons of methylation and structural variants from low-input whole-genome methylation sequencing.
Zhifu SunSaurabh BehatiPanwen WangAditya BhagwateSamantha McDonoughVivian WangWilliam TaylorJulie CunninghamJohn KisielPublished in: Epigenomics (2023)
Aim: Whole-genome methylation sequencing carries both DNA methylation and structural variant information (single nucleotide variant [SNV]; copy number variant [CNV]); however, limited data is available on the reliability of obtaining this information simultaneously from low-input DNA using various library preparation and sequencing protocols. Methods: A HapMap NA12878 sample was sequenced with three protocols (EM-sequencing, QIA-sequencing and Swift-sequencing) and their performance was compared on CpG methylation measurement and SNV and CNV detection. Results: At low DNA input (10-25 ng), EM-sequencing was superior in almost all metrics except CNV detection where all protocols were similar. EM-sequencing captured the highest number of CpGs and true SNVs. Conclusion: EM-sequencing is suitable to detect methylation, SNVs and CNVs from single sequencing with low-input DNA.