Quantifying Pathogen Surveillance Using Temporal Genomic Data

Chan, Joseph M.; Rabadan, Raul

The q2 coefficient of West Nile virus in the United States, on the other hand, was higher in the first few years after its introduction into the United States (see Fig. S4 and Table S5 in the supplemental material). However, despite the implementation of early warning systems for West Nile virus, by sampling dead birds (33) and mosquito populations (34), the q2 coefficient beyond 2003 dropped over time, even with the addition of multiple isolates within a short period of time. This decline in the q2 coefficient coincided with a rapid population growth of West Nile virus in the United States during 2002 and 2003 spurred by the establishment of the WN02 strain marked by a V159A mutation in its E protein (35) . This analysis suggests that current surveillance is being outpaced by the growing diversity of the virus in the New World. Other genetic distances besides the Hamming distance produced a Ͻ0.094 change in the q2 coefficient of West Nile virus.