Showing papers by "Edward Allen Wenger published in 2018"

Implementation and applications of EMOD, an individual-based multi-disease modeling platform

Abstract: Individual-based models provide modularity and structural flexibility necessary for modeling of infectious diseases at the within-host and population levels, but are challenging to implement. Levels of complexity can exceed the capacity and timescales for students and trainees in most academic institutions. Here we describe the process and advantages of a multi-disease framework approach developed with formal software support. The epidemiological modeling software, EMOD, has undergone a decade of software development. It is structured so that a majority of code is shared across disease modeling including malaria, HIV, tuberculosis, dengue, polio and typhoid. In additional to implementation efficiency, the sharing increases code usage and testing. The freely available codebase also includes hundreds of regression tests, scientific feature tests and component tests to help verify functionality and avoid inadvertent changes to functionality during future development. Here we describe the levels of detail, flexible configurability and modularity enabled by EMOD and the role of software development principles and processes in its development.

Seasonality and heterogeneity of malaria transmission determine success of interventions in high-endemic settings: a modeling study.

Abstract: Malaria transmission is both seasonal and heterogeneous, and mathematical models that seek to predict the effects of possible intervention strategies should accurately capture realistic seasonality of vector abundance, seasonal dynamics of within-host effects, and heterogeneity of exposure, which may also vary seasonally. Prevalence, incidence, asexual parasite and gametocyte densities, and infectiousness measurements from eight study sites in sub-Saharan Africa were used to calibrate an individual-based model with innate and adaptive immunity. Data from the Garki Project was used to fit exposure rates and parasite densities with month-resolution. A model capturing Garki seasonality and seasonal heterogeneity of exposure was used as a framework for characterizing the infectious reservoir of malaria, testing optimal timing of indoor residual spraying, and comparing four possible mass drug campaign implementations for malaria control. Seasonality as observed in Garki sites is neither sinusoidal nor box-like, and substantial heterogeneity in exposure arises from dry-season biting. Individuals with dry-season exposure likely account for the bulk of the infectious reservoir during the dry season even when they are a minority in the overall population. Spray campaigns offer the most benefit in prevalence reduction when implemented just prior to peak vector abundance, which may occur as late as a couple months into the wet season, and targeting spraying to homes of individuals with dry-season exposure can be particularly effective. Expanding seasonal malaria chemoprevention programs to cover older children is predicted to increase the number of cases averted per treatment and is therefore recommended for settings of seasonal and intense transmission. Accounting for heterogeneity and seasonality in malaria transmission is critical for understanding transmission dynamics and predicting optimal timing and targeting of control and elimination interventions.

Modeling the genetic relatedness of Plasmodium falciparum parasites following meiotic recombination and cotransmission.

Abstract: Unlike in most pathogens, multiple-strain (polygenomic) infections of P. falciparum are frequently composed of genetic siblings. These genetic siblings are the result of sexual reproduction and can coinfect the same host when cotransmitted by the same mosquito. The degree with which coinfecting strains are related varies among infections and populations. Because sexual recombination occurs within the mosquito, the relatedness of cotransmitted strains could depend on transmission dynamics, but little is actually known of the factors that influence the relatedness of cotransmitted strains. Part of the uncertainty stems from an incomplete understanding of how within-host and within-vector dynamics affect cotransmission. Cotransmission is difficult to examine experimentally but can be explored using a computational model. We developed a malaria transmission model that simulates sexual reproduction in order to understand what determines the relatedness of cotransmitted strains. This study highlights how the relatedness of cotransmitted strains depends on both within-host and within-vector dynamics including the complexity of infection. We also used our transmission model to analyze the genetic relatedness of polygenomic infections following a series of multiple transmission events and examined the effects of superinfection. Understanding the factors that influence the relatedness of cotransmitted strains could lead to a better understanding of the population-genetic correlates of transmission and therefore be important for public health.

Modeling the impact of Plasmodium falciparum sexual stage immunity on the composition and dynamics of the human infectious reservoir for malaria in natural settings.

Abstract: Malaria transmission remains high in Sub-Saharan Africa despite large-scale implementation of malaria control interventions. A comprehensive understanding of the transmissibility of infections to mosquitoes may guide the design of more effective transmission reducing strategies. The impact of P. falciparum sexual stage immunity on the infectious reservoir for malaria has never been studied in natural settings. Repeated measurements were carried out at start-wet, peak-wet and dry season, and provided data on antibody responses against gametocyte/gamete antigens Pfs48/45 and Pfs230 as anti-gametocyte immunity. Data on high and low-density infections and their infectiousness to anopheline mosquitoes were obtained using quantitative molecular methods and mosquito feeding assays, respectively. An event-driven model for P. falciparum sexual stage immunity was developed and fit to data using an agent based malaria model infrastructure. We found that Pfs48/45 and Pfs230 antibody densities increased with increasing concurrent gametocyte densities; associated with 55-70% reduction in oocyst intensity and achieved up to 44% reduction in proportions of infected mosquitoes. We showed that P. falciparum sexual stage immunity significantly reduces transmission of microscopic (p < 0.001) but not submicroscopic (p = 0.937) gametocyte infections to mosquitoes and that incorporating sexual stage immunity into mathematical models had a considerable impact on the contribution of different age groups to the infectious reservoir of malaria. Human antibody responses to gametocyte antigens are likely to be dependent on recent and concurrent high-density gametocyte exposure and have a pronounced impact on the likelihood of onward transmission of microscopic gametocyte densities compared to low density infections. Our mathematical simulations indicate that anti-gametocyte immunity is an important factor for predicting and understanding the composition and dynamics of the human infectious reservoir for malaria.

Local and regional dynamics of chikungunya virus transmission in Colombia: the role of mismatched spatial heterogeneity.

Abstract: Mathematical models of transmission dynamics are routinely fitted to epidemiological time series, which must inevitably be aggregated at some spatial scale. Weekly case reports of chikungunya have been made available nationally for numerous countries in the Western Hemisphere since late 2013, and numerous models have made use of this data set for forecasting and inferential purposes. Motivated by an abundance of literature suggesting that the transmission of this mosquito-borne pathogen is localized at scales much finer than nationally, we fitted models at three different spatial scales to weekly case reports from Colombia to explore limitations of analyses of nationally aggregated time series data. We adapted the recently developed Disease Transmission Kernel (DTK)-Dengue model for modeling chikungunya virus (CHIKV) transmission, given the numerous similarities of these viruses vectored by a common mosquito vector. We fitted versions of this model specified at different spatial scales to weekly case reports aggregated at different spatial scales: (1) single-patch national model fitted to national data; (2) single-patch departmental models fitted to departmental data; and (3) multi-patch departmental models fitted to departmental data, where the multiple patches refer to municipalities within a department. We compared the consistency of simulations from fitted models with empirical data. We found that model consistency with epidemic dynamics improved with increasing spatial granularity of the model. Specifically, the sum of single-patch departmental model fits better captured national-level temporal patterns than did a single-patch national model. Likewise, multi-patch departmental model fits better captured department-level temporal patterns than did single-patch departmental model fits. Furthermore, inferences about municipal-level incidence based on multi-patch departmental models fitted to department-level data were positively correlated with municipal-level data that were withheld from model fitting. Our model performed better when posed at finer spatial scales, due to better matching between human populations with locally relevant risk. Confronting spatially aggregated models with spatially aggregated data imposes a serious structural constraint on model behavior by averaging over epidemiologically meaningful spatial variation in drivers of transmission, impairing the ability of models to reproduce empirical patterns.

DTK-Dengue: A new agent-based model of dengue virus transmission dynamics

Abstract: Dengue virus (DENV) is a pathogen spread by Aedes mosquitoes that has a considerable impact on global health. Agent-based models can be used to represent explicitly represent factors that are difficult to measure empirically, by focusing on specific aspects of DENV transmission dynamics that influence spread in a particular location. We present a new agent-based model for DENV dynamics, DTK-Dengue, that can be readily applied to new locations and to a diverse set of goals. It extends the vector-borne disease module in the Institute for Disease Modelling's Epidemiological Modeling Disease Transmission Kernel (EMOD-DTK) to model DENV dynamics. There are three key modifications present in DTK-Dengue: 1) modifications to how climatic variables influence vector development for Aedes mosquitoes, 2) updates to adult vector behavior to make them more similar to Aedes, and 3) the inclusion of four DENV serotypes, including their effects on human immunity and symptoms. We demonstrate DTK-Dengue's capabilities by fitting the model to four interrelated datasets: total and serotype-specific dengue incidences between January 2007 and December 2008 from San Juan, Puerto Rico; the age distribution of reported dengue cases in Puerto Rico during 2007; and the number of adult female Ae. aegypti trapped in two neighborhoods of San Juan between November 2007 and December 2008. The model replicated broad patterns in the reference data, including a correlation between vector population dynamics and rainfall, appropriate seasonality in the reported incidence, greater circulation of DENV-3 than any other serotype, and an inverse relationship between age and the proportion of cases associated with each age group over 20 years old. This exercise demonstrates the potential for DTK-Dengue to assimilate multiple types of epidemiologic data into a realistic portrayal of DENV transmission dynamics. Due to the open availability of the DTK-Dengue software and the availability of numerous other modules for modeling disease transmission and control from EMOD-DTK, this new model has potential for a diverse range of future applications in a wide variety of settings.