A comparative study of ultra-deep pyrosequencing and cloning to quantitatively analyze the viral quasispecies using hepatitis B virus infection as a model

• Cloning has been the reference method to study viral quasispecies composition.
• UDPS yields reproducible, quantitative results similar to in vitro techniques.
• Haplotypes with deletions and insertions are only detected by cloning.

In this study, the reliability and reproducibility of viral quasispecies quantification by three ultra-deep pyrosequencing (UDPS) methods (FLX+, FLX, and Junior) were investigated and results compared with the conventional cloning technique. Hepatitis B virus (HBV) infection was selected as the model. The preCore/Core region, the least overlapped HBV region, was analyzed in samples from a chronic hepatitis B patient by cloning and by UDPS.After computation filtering of the UDPS results, samples A1 and A2 (FLX+) and sample B (FLX) yielded the same 20 polymorphic positions. Junior yielded 18 polymorphic positions that coincided with the FLX results. In contrast, 50 polymorphic positions were detected by cloning. Quasispecies complexity plotted on graphs showed superimposed patterns and the quantitative parameters were similar between FLX+, FLX, Junior, and the cloning sequences. Twenty-two haplotypes were detected by Junior, and 37, 40, and 39 were detected by FLX A1, A2, and B, respectively. These differences may be attributable to methodological differences between FLX and Junior. By cloning, 47 haplotypes were detected. Eight clones with insertions and deletions that induced de novo stop codons were not observed by UDPS because the UDPS filter discarded them.Our results indicate that UDPS is an optimal alternative to molecular cloning for quantitative study of the viral quasispecies. Nonetheless, specific mutations, such as insertions and deletions, were only detected by cloning. A filter should be designed to analyze cloning sequences, and UDPS filters should be improved to include the specific mutations.