Extended Data Fig. 4: Cross-comparison of sequencing approaches.
From: SARS-CoV-2 evolution during treatment of chronic infection
a, Comparison between short-read (Illumina) and long-read single-molecule (Oxford Nanopore) sequencing methods for the six observed mutations in the spike protein. Concordance was generally good between the majority of time points; however, owing to large discrepancies in a number of time points, we suggest that due to the high base-calling error rate, the Nanopore method is not yet suitable for calling minority variants. As such, all figures in the main paper were produced using only the Illumina data. b, Single-genome sequencing data from respiratory samples at the indicated days. The number of single genomes obtained at each time point with the mutations of interest (identified by deep sequencing) are shown. *The denominator is 19, as the primer reads for two samples were poor quality at amino acid 796 on day 98. Amino acid variant and corresponding nucleotide position: spike(W64G), 21752; spike(Δ69), 21765–21770; spike(Y200H), 22160; spike(T240I), 22281; spike(P330S), 22550; spike(D795H), 23948.
