Targeted Sanger sequencing to recover key mutations in SARS-CoV-2 variant genome assemblies produced by next-generation sequencing.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is adaptively evolving to ensure its persistence within human hosts. It is therefore necessary to continuously monitor the emergence and prevalence of novel variants that arise. Importantly, some mutations have been associated with both molecular diagnostic failures and reduced or abrogated next-generation sequencing (NGS) read coverage in some genomic regions. Such impacts are particularly problematic when they occur in genomic regions such as those that encode the spike (S) protein, which are crucial for identifying and tracking the prevalence and dissemination dynamics of concerning viral variants. Targeted Sanger sequencing presents a fast and cost-effective means to accurately extend the coverage of whole-genome sequences. We designed a custom set of primers to amplify a 401 bp segment of the receptor-binding domain (RBD) (between positions 22698 and 23098 relative to the Wuhan-Hu-1 reference). We then designed a Sanger sequencing wet-laboratory protocol. We applied the primer set and wet-laboratory protocol to sequence 222 samples that were missing positions with key mutations K417N, E484K, and N501Y due to poor coverage after NGS sequencing. Finally, we developed SeqPatcher, a Python-based computational tool to analyse the trace files yielded by Sanger sequencing to generate consensus sequences, or take preanalysed consensus sequences in fasta format, and merge them with their corresponding whole-genome assemblies. We successfully sequenced 153 samples of 222 (69 %) using Sanger sequencing and confirmed the occurrence of key beta variant mutations (K417N, E484K, N501Y) in the S genes of 142 of 153 (93 %) samples. Additionally, one sample had the Y508F mutation and four samples the S477N. Samples with RT-PCR Ct scores ranging from 13.85 to 37.47 (mean=25.70) could be Sanger sequenced efficiently. These results show that our method and pipeline can be used to improve the quality of whole-genome assemblies produced using NGS and can be used with any pairs of the most used NGS and Sanger sequencing platforms.