Large-scale evolutionary surveillance of the 2009 H1N1 influenza A virus using resequencing arrays

Lee, Charlie Wah Heng; Koh, Chee Wee; Chan, Yang Sun; Aw, Pauline Poh Kim; Loh, Kuan Hon; Han, Bing Ling; Thien, Pei Ling; Nai, Geraldine Yi Wen; Hibberd, Martin L.; Wong, Christopher W.; Sung, Wing-Kin

Unlike the NHIPs of wildtype and true-mutation calls, the NHIPs of most errors and 'N' calls appear haphazard ( Figure 5 ). However, when we traced the locations of these errors and 'N' calls on the genome, we found that some are isolated among good calls while others are conjugated in a small locality of the genome. We investigated the NHIPs of isolated errors and 'N' calls that occurred among good calls and found that in these NHIPs, only the PM probe of the query base that is an error or 'N' call has poor hybridization differentiation with its MM probes while other PM probes have hybridization intensities significantly higher than that of their MM probes in general ( Figure 6 ). This suggests that for such calls, only the PM and MM probes of the query base are noisy while neighbouring PM and MM probes are unaffected. In addition, we also found that long chains of consecutive error and 'N' calls (especially at the 5 0 -and 3 0 -end of the sample sequences) often have NHIPs where the PM probe of the query base together with neighbouring PM probes, have poor hybridization differentiation with their MM probes (Figure 7) . These error and 'N' calls usually occur at the ends of the genome segments. In summary, NHIP analysis showed that all true mutation calls had a characteristic profile (Figure 2b ) that differed from wild-type sequence calls (Figure 2a ). Ambiguous calls arising from different causes, such as homopolymers, isolated errors and hybridization artifacts also have profiles that are distinct from true mutation profiles ( Figure 2 ). See 'Materials and Methods' section for details.