Benchmarking of Oxford Nanopore whole genome sequencing for germline variant and CpG methylation detection across Canada's national platform for genome sequencing and analysis.

Advances in sequencing technology have enabled population-level Whole Genome Sequencing (WGS) efforts to be undertaken in many countries. Often, this requires collaboration across a distributed network of sequencing centres to allow efficient use of existing resources. Previously we tested the robustness of short-read sequencing technology and analysis pipelines across three established sequencing centres located in Montreal, Toronto, and Vancouver, constituting CGEn, Canada's national platform for genome sequencing and analysis (www.cgen.ca). In this work, we extend the study to cover Oxford Nanopore Technologies (ONT) long read-based WGS technology which is increasingly being used for large-scale genomics studies. Thus, we performed ONT WGS of the HG002 cell line, a well-characterized standard obtained directly from the Coriell Institute, aiming for a minimum of 30× coverage using one R10.4 PromethION flowcell at each centre. The sequencing datasets were analyzed using commonly developed pipelines for SNVs, Indels, SV, and CpG methylation detection and then compared to the relevant publicly available GIAB benchmark datasets. As a result, we tested the robustness of the laboratory protocols as well as the effectiveness of the analytical pipelines for simultaneous analysis of genomic variation and CpG methylation. Key findings include: SNV detection with higher F1-scores in RefSeq Coding regions for the ONT datasets (99.1%-99.5%) compared to Illumina NovaSeqX data (96.5%); additionally, there was high correlation of CpG methylation across all the sequencing centres (R = 0.97), as well as with publicly available WGBS (R = 0.88) and EM-Seq (R = 0.93) data from the EpiQC study.