Quantifying Pathogen Surveillance Using Temporal Genomic Data

Chan, Joseph M.; Rabadan, Raul

We compared the q2 coefficient to other possible quantitative methods, including those based on phylogenetics and clustering. A drawback common to many of these surveillance methods is that they are slow, computationally expensive techniques that are not explicitly designed to capture pathogen diversity. For this reason, we developed the q2 coefficient, which is readily computable and captures surveillance at any given time point. Applying the q2 coefficient revealed deficient sampling of swine H1N1 and nonavian H5N1, dengue, and West Nile viruses. These results bear great potential to guide the future allocation of energy and resources for gathering viral isolates. We also foresee further ecological applications of the q2 coefficient as an effective measure of asymptotic species richness without relying on arbitrary criteria for defining species, unlike classic techniques like Chao's estimate. This feature can be particularly valuable for assessing the number of additional samples needed to represent all of the species of an organism in metagenomic studies. In conclusion, the scientific community as a whole must improve its surveillance networks and share information for the advancement of scientific inquiry and public health interventions, and we offer the q2 coefficient as a means of evaluating the effectiveness of such efforts.