Bioinformatic investigation of discordant sequence data for SARS-CoV-2: insights for robust genomic analysis during pandemic surveillance

Abstract

The capacity to undertake whole genome sequencing (WGS) in public health laboratories (PHLs) has grown rapidly in response to COVID-19, and SARS-CoV-2 genomic data has been invaluable for managing the pandemic. The public health response has been further supported by the rapid upgrade and implementation of laboratory and bioinformatic resources. However, there remains a high degree of variability in methods and capabilities between laboratories. In addition to evolving methodology and improved understanding of SARS-CoV-2, public health laboratories have become strained during surges in case numbers, adding to the difficulty of ensuring the highest data accuracy. Here, we formed a national working group comprised of laboratory scientists and bioinformaticians from Australia and New Zealand to improve data concordance across PHLs. Through investigating discordant sequence data from Australia's first external SARS-CoV-2 WGS proficiency testing program (PTP), we show that most discrepancies in genome assessment arose from intrahost variation. While others could be remedied using reasonable, parsimonious bioinformatic quality control. Furthermore, we demonstrate how multidisciplinary national working groups can inform guidelines in real time for bioinformatic quality acceptance criteria. Provision of technical feedback allows laboratory improvement during a pandemic in real time, enhancing public health responses.