When a new virus emerges, it’s important to understand how it might spread through a population so protocols can be put in place to help contain the spread. Dr. Cécile Tran-Kiem, a postdoctoral researcher in Dr. Trevor Bedford’s lab in the Vaccine and Infectious Disease Division, explains that there are several factors that can influence the extent to which a virus or other pathogen, might spread. These factors include “transmission intensity (that can be measured by the mean number of persons infected by a case, also called the reproduction number) and transmission heterogeneity (how that number varies between cases). Quantifying transmission intensity is important to understand whether a pathogen can spread in a population,” and multiple methods are available to calculate this. However, measuring transmission heterogeneity has proved more of a challenge. If heterogeneity is high across a pathogen group, “a small number of individuals plays a disproportionate role to the spread of a pathogen and targeting control measures towards those individuals can be very effective to reduce the epidemic burden. On the other hand, if heterogeneity is low, such targeted measures wouldn’t be as valuable and population-level interventions might be more relevant,” says Dr. Tran-Kiem. In a recently published PNAS study, Dr. Tran-Kiem “developed a new method to estimate these two key parameters from pathogen genomic data” to better predict the spread of pathogens.
The size of an epidemiological cluster, or the number of infected people, is expected to be impacted by transmission intensity. Thus, previous approaches have used epidemiological cluster size to characterize transmission heterogeneity and intensity. For example, if each case of transmission only “infects a small number of persons, transmission chains will quickly die out so that we will only observe small epidemiological clusters. However, if each case infects a greater number of individuals, transmission chains will circulate for longer, and these clusters will tend to be larger. In a similar manner, transmission heterogeneity will also shape cluster sizes,” Dr. Tran-Kiem explains. Thus, “there is a direct relationship between the size distribution of epidemiological clusters and transmission intensity and heterogeneity,” she adds. Consistent with this idea, Dr. Tran-Kiem notes, “when we looked at the size of clusters of identical sequences by variant, we saw that this was reflecting really well this serial replacement of less transmissible variants by more transmissible ones.”
Taking this observation of how transmission characteristics can shape cluster size, the Bedford lab developed a tool to assess how transmissible a specific variant is based on the size distribution of identical genome sequences. Dr. Tran-Kiem explains that the method they developed “was 2-fold: 1) we provide the tools to estimate transmission intensity and heterogeneity from identical cluster size distribution and 2) we provide the tools to characterize the transmission advantage of a variant from the same cluster size distribution.” While this method entails some complex math, its basis lies in taking into account the pathogen reproduction number, R, and the dispersion parameter, k. Compared to other approaches that also use pathogen genome sequences to characterize transmission, this approach “does not require inferring a phylogenetic tree (which can quickly become computationally very expensive),” says Dr. Tran-Kiem. Additionally, not utilizing a phylogenetic tree makes this method “easily scalable to larger pathogen genome datasets such as those generated during the pandemic.”
After developing this tool, Dr. Tran-Kiem applied it to various cases of pathogen spread, including evaluating measles outbreaks in the post-vaccination era and SARS-CoV-2 transmission heterogeneity in New Zealand during a period when the epidemic was mostly suppressed without reported community transmission. Their results were in line with previous analyses. Additionally, the authors applied their method to SARS-CoV-2 variants in Washington State, where they were “able to detect changes in the transmissibility of known variants of concern (Alpha, Delta, Omicron) from the distribution of identical cluster sizes,” says Tran-Kiem. Dr. Tran-Kiem acknowledges that one limitation of their method, is that it is “only valid in settings with lower transmission intensity. It would be interesting to extend this work to situations where a pathogen would spread more quickly. It would also be interesting to explore whether we can use the size of clusters of identical sequences to understand how transmission might vary spatially or between demographic groups.” While there is room to expand this method to assess additional pathogen-spreading characteristics, the current method provides a much-needed tool to better characterize pathogens by looking at population-level parameters, which can inform epidemic control.
This work was supported by the Howard Hughes Medical Institute and the National Institutes of Health.
Tran-Kiem C, Bedford T. Estimating the reproduction number and transmission heterogeneity from the size distribution of clusters of identical pathogen sequences. Proc Natl Acad Sci. 2024.
