LAUSR

In the last 3 years, the COVID-19 pandemic has touched every corner of the globe, resulting in substantial mortality, a number of chronic health problems, and long-lasting economic repercussions. The pandemic also had a profound effect on how scientific data are reported and disseminated, as researchers rushed to share their findings and discovered new ways to collaborate despite the challenges posed by lockdowns and other disruptions. Throughout this period, Genome Biology and Evolution welcomed papers that deepen our knowledge of SARS-CoV-2, the virus that causes COVID-19. The journal’s Virtual Issue on SARS-CoV-2 collates some of the key papers published in GBE that have helped to shed light on the origins and evolution of this world-changing virus. The first in the Virtual Issue is a recent study by Petrone et al. (2023) that analyzed differences in reporting delays among SARS-CoV-2 variants of concern (VOCs). The authors found that the reporting of genomic data for the Mu variant was significantly slower than that of other VOCs and that this trend could be observed across different countries, states, and even individual laboratories. Perhaps due in part to this reporting bias, Mu was not designated as a VOC until approximately 1 year after it emerged, which is substantially longer than the Delta and Omicron VOCs. According to Mary E. Petrone, a lead author on the study who has since completed her PhD from Yale University and is now at the University of Sydney, the researchers originally noticed this bias when reporting their own genomic data for SARS-CoV-2 to the Global Initiative on Sharing Avian Influenza Data (GISAID), which expanded their database to include SARS-CoV-2. “We noticed that a large number of our submissions to GISAID were getting rejected due to perceived quality control issues. When we took a closer look, we realized that a substantial portion of the rejected sequences were the Mu variant.” This may be because Mu carries a four-nucleotide deletion in ORF3a that causes a frameshift mutation that may often be flagged as a sequencing error. Fortunately, Mu did not spread globally to the same extent as Delta and Omicron. However, Petrone notes that such reporting biases must be addressed in order to avoid a delayed response to a VOC in the future: “We hope that our study will serve as a motivating factor to reduce genomic surveillance gaps for SARS-CoV-2 and other viruses of public health importance.” Another recent article comes from Aiewsakun et al. (2023), who investigated the adaptation of SARS-CoV-2 to three cell lines that are commonly used to grow viral stocks for laboratory studies of the virus. They identified “numerous potential cell culture adaptation changes scattered across the entire virus genome.” In order to avoid confounding effects of these changes on lab experiments, the authors emphasize that it is important to perform deepsequencing of cultured viral samples when performing sensitive SARS-CoV-2 experiments. They also noted that, among the three cell lines tested, Vero E6/TMPRSS2 cells induced the fewest changes and thus might be the best option for SARS-CoV-2 culture. In addition to cell culture studies, animal models of SARS-CoV-2 infection and COVID-19 disease are critical for understanding how to prevent and treat the worst outcomes. Whereas the Syrian hamster is widely used as a model for mild-to-moderate COVID-19, the Roborovski dwarf hamster has become an important model for severe-to-lethal COVID-19. Although this model was previously limited by the absence of a full genome annotation, Andreotti et al. (2022) have provided a draft genome assembly that includes gene annotations and that “can now serve as a key resource for many research projects employing this hamster species.” Two papers in the Virtual Issue investigate SARS-CoV-2 genotypes and polymorphisms in order to reveal patterns of viral diversity. Moustafa and Planet (2021) developed a tool called GNUVID (GNU-based Virus Identification) that integrates whole-genome multilocus sequence typing and machine learning and used it to assign all high-quality SARS-CoV-2 genomes available from GISAID to sequence type profiles. Their results revealed introduction and