Assembly deconvolution resolves SARS-CoV-2 haplotypes in Delta-Omicron co-infections

Abstract

SARS-CoV-2 baseline surveillance requires read-mapping onto a reference, yielding individual assemblies. This process obscures any intra-host viral sequence heterogeneity. Delta/Omicron strains co-circulated in Los Angeles during the winter of 2021. We sequenced two samples referred to us from our clinical laboratory partners. Their assemblies exhibited ambiguities indicative of co-infection. We screened loci of aligned read datasets to rapidly identify and/or rule out suspected Delta-Omicron co-infections. We developed a reference-based assembly deconvolution method to resolve each consensus sequence from suspected co-infecting viruses. Our experience illustrates the benefit of having raw sequencing data and/or aligned reads available in addition to genome assemblies.