Quantifying Pathogen Surveillance Using Temporal Genomic Data

Chan, Joseph M.; Rabadan, Raul

Influenza virus surveillance. Complete coding sequences of hemagglutinin (HA), the major antigenic segment of influenza virus, were collected from the NCBI (National Center for Biotech-nology Information) and GISAID databases and multiply aligned. To compare the levels of surveillance of different strains, we separately considered sequences for human (H3N2, seasonal H1N1 pre-2009, SOIV H1N1 post-2009, and H5N1), avian (H5N1 and non-H5N1), and swine H1N1 influenza viruses. Figure 1A and B depict the absolute number of sequences isolated across time as a first measure of surveillance. However, we wanted a measure of surveillance that would take into account the evolutionary rate of the virus, as well as the biodiversity of the viral pool. Calculation of Shannon's entropy in Fig. 1C is an effective measure of genetic diversity but, by itself, is no measure of surveillance.