Inferring pandemic growth rates from sequence data

de Silva, Eric; Ferguson, Neil M.; Fraser, Christophe

As an example, figure 1 shows a retrospective BSP generated using BEAST (for temporal frequency, see the electronic supplementary material, figure S1) of the effective population size of 2009 H1N1 derived from the 110 publically available complete haemagglutinin (HA) viral sequences (excluding duplicates) listed in the NCBI influenza virus resource [10] as arising from specimens collected in the USA over the period 7 April 2009 to 24 May 2009. This analysis covers a seven week period at the start of the pandemic, and thus extends the initial analyses presented in Fraser et al. [8] . Figure 1 shows a slowdown in the growth rate of the effective population size around late April/early May. If this reflects slowing growth in the number of infected individuals, then this would be of epidemiological interest-perhaps suggesting some impact of public health measures enacted in the USA at that time.