The spatial dynamics of arbovirus transmission are a complex function of the pathogen, the environment, vector populations, and humans. Dengue (DENV), a mosquito-borne arbovirus, has seen a rapid global spread and a dramatic growth in incidence with an estimated 390 million infections per year. In many cases, the drivers of this spread are unclear. On fine spatial scales, human movement defines spatial patterns of dengue incidence and exposure risk. Predicting the spatial and temporal patterns of dengue transmission therefore requires quantifying human mobility patterns. Here, we will focus on DENV in Sri Lanka as a case study for the integration of multiple sources of human mobility data with epidemiologic and phylogenetic data. In the proposed study, we will use a wealth of existing data and will collect unique, linked mobility and genomic data from the primary DHF referral hospital in Sri Lanka to predict the spatial dynamics from the human population (travel data) and virus (viral genomics). Through a prospective study based at a referral hospital in Negombo, Sri Lanka, we will collect mobility (travel survey and GPS loggers) and viral genomes from DENV positive patients. These data will be used to characterize individual travel patterns that will be used to statistically identify the risk of DENV infection at different locations. We will compare these results with population-level mobility patterns from mobile phone calling records in a model of DENV informed by national incidence data. We will pair this analysis with a phylogenetic analysis of whole genome sequenced DENV viral genomes to identify spatial transmission chains and the geographic relatedness between transmission pairs. Using the genomic information, we will identify the utility of travel and epidemiological data to risk factors related to spatial transmission.

Dengue (DENV), a mosquito-borne flavivirus, infects approximately 390 million people globally each year and there is a clear need for a better understanding to optimally allocate vector control, surveillance efforts, and available clinical care. However, current surveillance programs based on routine reporting from health systems cannot identify chains of transmission or locate fine-scale risk of exposure. Using dengue virus (DENV) transmission in Sri Lanka as a case study, we will investigate the utility of integrating human travel and pathogen genomic information to inform predictive risk models of spatial DENV transmission.