Spatio-temporal spillover risk of yellow fever in Brazil

BACKGROUND

Yellow fever virus is a mosquito-borne flavivirus that persists in an enzoonotic cycle in non-human primates (NHPs) in Brazil, causing disease in humans through spillover events. Yellow fever (YF) re-emerged in the early 2000s, spreading from the Amazon River basin towards the previously considered low-risk, southeastern region of the country. Previous methods mapping YF spillover risk do not incorporate the temporal dynamics and ecological context of the disease, and are therefore unable to predict seasonality in spatial risk across Brazil. We present the results of a bagged logistic regression predicting the propensity for YF spillover per municipality (administrative sub-district) in Brazil from environmental and demographic covariates aggregated by month. Ecological context was incorporated by creating National and Regional models of spillover dynamics, where the Regional model consisted of two separate models determined by the regions' NHP reservoir species richness (high vs low).

RESULTS

Of the 5560 municipalities, 82 reported YF cases from 2001 to 2013. Model accuracy was high for the National and low reservoir richness (LRR) models (AUC = 0.80), while the high reservoir richness (HRR) model accuracy was lower (AUC = 0.63). The National model predicted consistently high spillover risk in the Amazon, while the Regional model predicted strong seasonality in spillover risk. Within the Regional model, seasonality of spillover risk in the HRR region was asynchronous to the LRR region. However, the observed seasonality of spillover risk in the LRR Regional model mirrored the national model predictions.

CONCLUSIONS

The predicted risk of YF spillover varies with space and time. Seasonal trends differ between regions indicating, at times, spillover risk can be higher in the urban coastal regions than the Amazon River basin which is counterintuitive based on current YF risk maps. Understanding the spatio-temporal patterns of YF spillover risk could better inform allocation of public health services.

Optimal treatment allocations in space and time for on-line control of an emerging infectious disease

A key component in controlling the spread of an epidemic is deciding where, when and to whom to apply an intervention. We develop a framework for using data to inform these decisions in realtime. We formalize a treatment allocation strategy as a sequence of functions, one per treatment period, that map up-to-date information on the spread of an infectious disease to a subset of locations where treatment should be allocated. An optimal allocation strategy optimizes some cumulative outcome, e.g. the number of uninfected locations, the geographic footprint of the disease or the cost of the epidemic. Estimation of an optimal allocation strategy for an emerging infectious disease is challenging because spatial proximity induces interference between locations, the number of possible allocations is exponential in the number of locations, and because disease dynamics and intervention effectiveness are unknown at out-break. We derive a Bayesian on-line estimator of the optimal allocation strategy that combines simulation-optimization with Thompson sampling. The estimator proposed performs favourably in simulation experiments. This work is motivated by and illustrated using data on the spread of white nose syndrome, which is a highly fatal infectious disease devastating bat populations in North America.

Carry-over effects of urban larval environments on the transmission potential of dengue-2 virus

Background

Mosquitoes are strongly influenced by environmental temperatures, both directly and indirectly via carry-over effects, a phenomenon by which adult phenotypes are shaped indirectly by the environmental conditions experienced in previous life stages. In landscapes with spatially varying microclimates, such as a city, the effects of environmental temperature can therefore lead to spatial patterns in disease dynamics. To explore the contribution of carry-over effects on the transmission of dengue-2 virus (DENV-2), we conducted a semi-field experiment comparing the demographic and transmission rates of Aedes albopictus reared on different urban land classes in the summer and autumn season. We parameterized a model of vectorial capacity using field- and literature-derived measurements to estimate the bias introduced into predictions of vectorial capacity not accounting for carry-over effects.

Results

The larval environment of different land classes and seasons significantly impacted mosquito life history traits. Larval development and survival rates were higher in the summer than the autumn, with no difference across land class. The effect of land class on adult body size differed across season, with suburban mosquitoes having the smallest wing length in the summer and the largest wing length in the autumn, when compared to other land classes. Infection and dissemination rates were higher in the autumn and on suburban and rural land classes compared to urban. Infectiousness did not differ across land class or season. We estimate that not accounting for carry-over effects can underestimate disease transmission potential in suburban and urban sites in the summer by up to 25%.

Conclusions

Our findings demonstrate the potential of the larval environment to differentially impact stages of DENV-2 infection in Ae. albopictus mosquitoes via carry-over effects. Failure to account for carry-over effects of the larval environment in mechanistic models can lead to biased estimates of disease transmission potential at fine-scales in urban environments.

Electronic supplementary material

The online version of this article (10.1186/s13071-018-3013-3) contains supplementary material, which is available to authorized users.

Anticipating epidemic transitions with imperfect data

Epidemic transitions are an important feature of infectious disease systems. As the transmissibility of a pathogen increases, the dynamics of disease spread shifts from limited stuttering chains of transmission to potentially large scale outbreaks. One proposed method to anticipate this transition are early-warning signals (EWS), summary statistics which undergo characteristic changes as the transition is approached. Although theoretically predicted, their mathematical basis does not take into account the nature of epidemiological data, which are typically aggregated into periodic case reports and subject to reporting error. The viability of EWS for epidemic transitions therefore remains uncertain. Here we demonstrate that most EWS can predict emergence even when calculated from imperfect data. We quantify performance using the area under the curve (AUC) statistic, a measure of how well an EWS distinguishes between numerical simulations of an emerging disease and one which is stationary. Values of the AUC statistic are compared across a range of different reporting scenarios. We find that different EWS respond to imperfect data differently. The mean, variance and first differenced variance all perform well unless reporting error is highly overdispersed. The autocorrelation, autocovariance and decay time perform well provided that the aggregation period of the data is larger than the serial interval and reporting error is not highly overdispersed. The coefficient of variation, skewness and kurtosis are found to be unreliable indicators of emergence. Overall, we find that seven of ten EWS considered perform well for most realistic reporting scenarios. We conclude that imperfect epidemiological data is not a barrier to using EWS for many potentially emerging diseases.

The description and number of undiscovered mammal species

Global species counts are a key measure of biodiversity and associated metrics of conservation. It is both scientifically and practically important to know how many species exist, how many undescribed species remain, and where they are found. We modify a model for the number of undescribed species using species description data and incorporating taxonomic information. We assume a Poisson distribution for the number of species described in an interval and use maximum likelihood to estimate parameter values of an unknown intensity function. To test the model's performance, we performed a simulation study comparing our method to a previous model under conditions qualitatively similar to those related to mammal species description over the last two centuries. Because our model more accurately estimates the total number of species, we predict that 5% of mammals remain undescribed. We applied our model to determine the biogeographic realms which hold these undescribed species.

Spatiotemporal Fluctuations and Triggers of Ebola Virus Spillover

Because the natural reservoir of Ebola virus remains unclear and disease outbreaks in humans have occurred only sporadically over a large region, forecasting when and where Ebola spillovers are most likely to occur constitutes a continuing and urgent public health challenge. We developed a statistical modeling approach that associates 37 human or great ape Ebola spillovers since 1982 with spatiotemporally dynamic covariates including vegetative cover, human population size, and absolute and relative rainfall over 3 decades across sub-Saharan Africa. Our model (area under the curve 0.80 on test data) shows that spillover intensity is highest during transitions between wet and dry seasons; overall, high seasonal intensity occurs over much of tropical Africa; and spillover intensity is greatest at high (>1,000/km²) and very low (<100/km²) human population densities compared with intermediate levels. These results suggest strong seasonality in Ebola spillover from wild reservoirs and indicate particular times and regions for targeted surveillance.

Estimating parasite host range

Estimating the number of host species that a parasite can infect (i.e. host range) provides key insights into the evolution of host specialism and is a central concept in disease ecology. Host range is rarely estimated in real systems, however, because variation in species relative abundance and the detection of rare species makes it challenging to confidently estimate host range. We applied a non-parametric richness indicator to estimate host range in simulated and empirical data, allowing us to assess the influence of sampling heterogeneity and data completeness. After validating our method on simulated data, we estimated parasite host range for a sparsely sampled global parasite occurrence database (Global Mammal Parasite Database) and a repeatedly sampled set of parasites of small mammals from New Mexico (Sevilleta Long Term Ecological Research Program). Estimation accuracy varied strongly with parasite taxonomy, number of parasite occurrence records, and the shape of host species-abundance distribution (i.e. the dominance and rareness of species in the host community). Our findings suggest that between 20% and 40% of parasite host ranges are currently unknown, highlighting a major gap in our understanding of parasite specificity, host-parasite network structure, and parasite burdens.

Experimental evidence of a pathogen invasion threshold

Host density thresholds to pathogen invasion separate regions of parameter space corresponding to endemic and disease-free states. The host density threshold is a central concept in theoretical epidemiology and a common target of human and wildlife disease control programmes, but there is mixed evidence supporting the existence of thresholds, especially in wildlife populations or for pathogens with complex transmission modes (e.g. environmental transmission). Here, we demonstrate the existence of a host density threshold for an environmentally transmitted pathogen by combining an epidemiological model with a microcosm experiment. Experimental epidemics consisted of replicate populations of naive crustacean zooplankton (Daphnia dentifera) hosts across a range of host densities (20-640 hosts l-1) that were exposed to an environmentally transmitted fungal pathogen (Metschnikowia bicuspidata). Epidemiological model simulations, parametrized independently of the experiment, qualitatively predicted experimental pathogen invasion thresholds. Variability in parameter estimates did not strongly influence outcomes, though systematic changes to key parameters have the potential to shift pathogen invasion thresholds. In summary, we provide one of the first clear experimental demonstrations of pathogen invasion thresholds in a replicated experimental system, and provide evidence that such thresholds may be predictable using independently constructed epidemiological models.

Comparative epidemiology of poliovirus transmission

Understanding the determinants of polio transmission and its large-scale epidemiology remains a public health priority. Despite a 99% reduction in annual wild poliovirus (WPV) cases since 1988, tackling the last 1% has proven difficult. We identified key covariates of geographical variation in polio transmission patterns by relating country-specific annual disease incidence to demographic, socio-economic and environmental factors. We assessed the relative contributions of these variables to the performance of computer-generated models for predicting polio transmission. We also examined the effect of spatial coupling on the polio extinction frequency in islands relative to larger land masses. Access to sanitation, population density, forest cover and routine vaccination coverage were the strongest predictors of polio incidence, however their relative effect sizes were inconsistent geographically. The effect of climate variables on polio incidence was negligible, indicating that a climate effect is not identifiable at the annual scale, suggesting a role for climate in shaping the transmission seasonality rather than intensity. We found polio fadeout frequency to depend on both population size and demography, which should therefore be considered in policies aimed at extinction. Our comparative epidemiological approach highlights the heterogeneity among polio transmission determinants. Recognition of this variation is important for the maintenance of population immunity in a post-polio era.

Residence time, native range size, and genome size predict naturalization among angiosperms introduced to Australia

Although critical to progress in understanding (i) if, and (ii) at what rate, introduced plants will naturalize and potentially become invasive, establishing causal links between traits and invasion success is complicated by data gaps, phylogenetic nonindependence of species, the inability to control for differences between species in residence time and propagule pressure, and covariance among traits. Here, we focus on statistical relationships between genomic factors, life history traits, native range size, and naturalization status of angiosperms introduced to Australia. In a series of analyses, we alternately investigate the role of phylogeny, incorporate introduction history, and use graphical models to explore the network of conditional probabilities linking traits and introduction history to naturalization status. Applying this ensemble of methods to the largest publicly available data set on plant introductions and their fates, we found that, overall, residence time and native range size best predicted probability of naturalization. Yet, importantly, probability of naturalization consistently increased as genome size decreased, even when the effects of shared ancestry and residence time in Australia were accounted for, and that this pattern was stronger in species without a history of cultivation, but present across annual-biennials, and herbaceous and woody perennials. Thus, despite introduction biases and indirect effects of traits via introduction history, across analyses, reduced genome size was nevertheless consistently associated with a tendency to naturalize.

Forecasting infectious disease emergence subject to seasonal forcing

BACKGROUND

Despite high vaccination coverage, many childhood infections pose a growing threat to human populations. Accurate disease forecasting would be of tremendous value to public health. Forecasting disease emergence using early warning signals (EWS) is possible in non-seasonal models of infectious diseases. Here, we assessed whether EWS also anticipate disease emergence in seasonal models.

METHODS

We simulated the dynamics of an immunizing infectious pathogen approaching the tipping point to disease endemicity. To explore the effect of seasonality on the reliability of early warning statistics, we varied the amplitude of fluctuations around the average transmission. We proposed and analyzed two new early warning signals based on the wavelet spectrum. We measured the reliability of the early warning signals depending on the strength of their trend preceding the tipping point and then calculated the Area Under the Curve (AUC) statistic.

RESULTS

Early warning signals were reliable when disease transmission was subject to seasonal forcing. Wavelet-based early warning signals were as reliable as other conventional early warning signals. We found that removing seasonal trends, prior to analysis, did not improve early warning statistics uniformly.

CONCLUSIONS

Early warning signals anticipate the onset of critical transitions for infectious diseases which are subject to seasonal forcing. Wavelet-based early warning statistics can also be used to forecast infectious disease.

Ultrasonic motor-induced geometric distortions in magnetic resonance images

Ultrasonic motors (USMs) are common actuators that can be safely used in the magnetic resonance imaging (MRI) environment. However, lack of MRI compatibility results in issues such as image distortion. This fact led researchers to shift focus from USMs to pneumatic and hydraulic actuators in development of surgical robots. The aim is to quantify and compensate the geometric distortion of MR images as generated by the presence of USMs. An ultrasonic motor was positioned in three orientations with respect to the bore axis. The induced distortions were compared across four image sequences. To reduce the distortions, three artifact reduction methods were employed. Geometric distortion is the only artifact in image slices farther from the motor. The various motor orientations lead to different distortions, with the lowest distortion for the z orientation. The maximum measured distortion of ten pixels occurred. This maximal distortion is equal to a 1-cm displacement of the displayed points relative to their actual locations and it is beyond the acceptable level for medical display standards. Bandwidth reduction reduced the distortion, with a 50% reduction for a doubled bandwidth. In conclusion, USMs can be preferred alternative because accurate targeting of pathologies can occur in free distorted images.

Anticipating the emergence of infectious diseases

In spite of medical breakthroughs, the emergence of pathogens continues to pose threats to both human and animal populations. We present candidate approaches for anticipating disease emergence prior to large-scale outbreaks. Through use of ideas from the theories of dynamical systems and stochastic processes we develop approaches which are not specific to a particular disease system or model, but instead have general applicability. The indicators of disease emergence detailed in this paper can be classified into two parallel approaches: a set of early-warning signals based around the theory of critical slowing down and a likelihood-based approach. To test the reliability of these two approaches we contrast theoretical predictions with simulated data. We find good support for our methods across a range of different model structures and parameter values.

Monitoring the Path to the Elimination of Infectious Diseases

During the endgame of elimination programs, parasite populations may exhibit dynamical phenomena not typical of endemic disease. Particularly, monitoring programs for tracking infection prevalence may be hampered by overall rarity, the sporadic and unpredictable timing and location of outbreaks, and under-reporting. A particularly important problem for monitoring is determining the distance that must be covered to achieve the elimination threshold at an effective reproduction number less than one. In this perspective, we suggest that this problem may be overcome by measuring critical slowing down. Critical slowing down is a phenomenon exhibited by nonlinear dynamical systems in the vicinity of a critical threshold. In infectious disease dynamics, critical slowing down is expressed as an increase in the coefficient of variation and other properties of the fluctuations in the number of cases. In simulations, we show the coefficient of variation to be insensitive to under-reporting error and therefore a robust measurement of the approach to elimination. Additionally, we show that there is an inevitable delay between the time at which the effective reproduction number is reduced to below one and complete elimination is achieved. We urge that monitoring programs include dynamical properties such as critical slowing down in their metrics for measuring achievement and avoid withdrawing control activities prematurely.

Predicting cryptic links in host-parasite networks

Networks are a way to represent interactions among one (e.g., social networks) or more (e.g., plant-pollinator networks) classes of nodes. The ability to predict likely, but unobserved, interactions has generated a great deal of interest, and is sometimes referred to as the link prediction problem. However, most studies of link prediction have focused on social networks, and have assumed a completely censused network. In biological networks, it is unlikely that all interactions are censused, and ignoring incomplete detection of interactions may lead to biased or incorrect conclusions. Previous attempts to predict network interactions have relied on known properties of network structure, making the approach sensitive to observation errors. This is an obvious shortcoming, as networks are dynamic, and sometimes not well sampled, leading to incomplete detection of links. Here, we develop an algorithm to predict missing links based on conditional probability estimation and associated, node-level features. We validate this algorithm on simulated data, and then apply it to a desert small mammal host-parasite network. Our approach achieves high accuracy on simulated and observed data, providing a simple method to accurately predict missing links in networks without relying on prior knowledge about network structure.

Experimental demonstration of an Allee effect in microbial populations

Microbial populations can be dispersal limited. However, microorganisms that successfully disperse into physiologically ideal environments are not guaranteed to establish. This observation contradicts the Baas-Becking tenet: 'Everything is everywhere, but the environment selects'. Allee effects, which manifest in the relationship between initial population density and probability of establishment, could explain this observation. Here, we experimentally demonstrate that small populations of Vibrio fischeri are subject to an intrinsic demographic Allee effect. Populations subjected to predation by the bacterivore Cafeteria roenbergensis display both intrinsic and extrinsic demographic Allee effects. The estimated critical threshold required to escape positive density-dependence is around 5, 20 or 90 cells ml(-1)under conditions of high carbon resources, low carbon resources or low carbon resources with predation, respectively. This work builds on the foundations of modern microbial ecology, demonstrating that mechanisms controlling macroorganisms apply to microorganisms, and provides a statistical method to detect Allee effects in data.

Density-dependent selection on mate search and evolution of Allee effects

Sexually reproducing organisms require males and females to find each other. Increased difficulty of females finding mates as male density declines is the most frequently reported mechanism of Allee effects in animals. Evolving more effective mate search may alleviate Allee effects, but may depend on density regimes a population experiences. In particular, high-density populations may evolve mechanisms that induce Allee effects which become detrimental when populations are reduced and maintained at a low density. We develop an individual-based, eco-genetic model to study how mating systems and fitness trade-offs interact with changes in population density to drive evolution of the rate at which males or females search for mates. Finite mate search rate triggers Allee effects in our model and we explore how these Allee effects respond to such evolution. We allow a population to adapt to several population density regimes and examine whether high-density populations are likely to reverse adaptations attained at low densities. We find density-dependent selection in most of scenarios, leading to search rates that result in lower Allee thresholds in populations kept at lower densities. This mainly occurs when fecundity costs are imposed on mate search, and provides an explanation for why Allee effects are often observed in anthropogenically rare species. Optimizing selection, where the attained trait value minimizes the Allee threshold independent of population density, depended on the trade-off between search and survival, combined with monogamy when females were searching. Other scenarios led to runaway selection on the mate search rate, including evolutionary suicide. Trade-offs involved in mate search may thus be crucial to determining how density influences the evolution of Allee effects. Previous studies did not examine evolution of a trait related to the strength of Allee effects under density variation. We emphasize the crucial role that mating systems, fitness trade-offs and the evolving sex have in determining the density threshold for population persistence, in particular since evolution need not always take the Allee threshold to its minimum value.

Accuracy of climate-based forecasts of pathogen spread

Species distribution models (SDMs) are a tool for predicting the eventual geographical range of an emerging pathogen. Most SDMs, however, rely on an assumption of equilibrium with the environment, which an emerging pathogen, by definition, has not reached. To determine if some SDM approaches work better than others for modelling the spread of emerging, non-equilibrium pathogens, we studied time-sensitive predictive performance of SDMs for Batrachochytrium dendrobatidis, a devastating infectious fungus of amphibians, using multiple methods trained on time-incremented subsets of the available data. We split our data into timeline-based training and testing sets, and evaluated models on each set using standard performance criteria, including AUC, kappa, false negative rate and the Boyce index. Of eight models examined, we found that boosted regression trees and random forests performed best, closely followed by MaxEnt. As expected, predictive performance generally improved with the length of time series used for model training. These results provide information on how quickly the potential extent of an emerging disease may be determined, and identify which modelling frameworks are likely to provide useful information during the early phases of pathogen expansion.

Accuracy of climate-based forecasts of pathogen spread

Species distribution models (SDMs) are a tool for predicting the eventual geographical range of an emerging pathogen. Most SDMs, however, rely on an assumption of equilibrium with the environment, which an emerging pathogen, by definition, has not reached. To determine if some SDM approaches work better than others for modelling the spread of emerging, non-equilibrium pathogens, we studied time-sensitive predictive performance of SDMs for Batrachochytrium dendrobatidis, a devastating infectious fungus of amphibians, using multiple methods trained on time-incremented subsets of the available data. We split our data into timeline-based training and testing sets, and evaluated models on each set using standard performance criteria, including AUC, kappa, false negative rate and the Boyce index. Of eight models examined, we found that boosted regression trees and random forests performed best, closely followed by MaxEnt. As expected, predictive performance generally improved with the length of time series used for model training. These results provide information on how quickly the potential extent of an emerging disease may be determined, and identify which modelling frameworks are likely to provide useful information during the early phases of pathogen expansion.

Data-driven identification of potential Zika virus vectors

Zika is an emerging virus whose rapid spread is of great public health concern. Knowledge about transmission remains incomplete, especially concerning potential transmission in geographic areas in which it has not yet been introduced. To identify unknown vectors of Zika, we developed a data-driven model linking vector species and the Zika virus via vector-virus trait combinations that confer a propensity toward associations in an ecological network connecting flaviviruses and their mosquito vectors. Our model predicts that thirty-five species may be able to transmit the virus, seven of which are found in the continental United States, including Culex quinquefasciatus and Cx. pipiens. We suggest that empirical studies prioritize these species to confirm predictions of vector competence, enabling the correct identification of populations at risk for transmission within the United States.

Do Biology Students Really Hate Math? Empirical Insights into Undergraduate Life Science Majors' Emotions about Mathematics

Undergraduate life science majors are reputed to have negative emotions toward mathematics, yet little empirical evidence supports this. We sought to compare emotions of majors in the life sciences versus other natural sciences and math. We adapted the Attitudes toward the Subject of Chemistry Inventory to create an Attitudes toward the Subject of Mathematics Inventory (ASMI). We collected data from 359 science and math majors at two research universities and conducted a series of statistical tests that indicated that four AMSI items comprised a reasonable measure of students' emotional satisfaction with math. We then compared life science and non-life science majors and found that major had a small to moderate relationship with students' responses. Gender also had a small relationship with students' responses, while students' race, ethnicity, and year in school had no observable relationship. Using latent profile analysis, we identified three groups-students who were emotionally satisfied with math, emotionally dissatisfied with math, and neutral. These results and the emotional satisfaction with math scale should be useful for identifying differences in other undergraduate populations, determining the malleability of undergraduates' emotional satisfaction with math, and testing effects of interventions aimed at improving life science majors' attitudes toward math.