Correction: Microbial phenotypic heterogeneity in response to a metabolic toxin: Continuous, dynamically shifting distribution of formaldehyde tolerance in Methylobacterium extorquens populations.

[This corrects the article DOI: 10.1371/journal.pgen.1008458.].

Quantifying the effectiveness of betaherpesvirus-vectored transmissible vaccines.

Transmissible vaccines have the potential to revolutionize how zoonotic pathogens are controlled within wildlife reservoirs. A key challenge that must be overcome is identifying viral vectors that can rapidly spread immunity through a reservoir population. Because they are broadly distributed taxonomically, species specific, and stable to genetic manipulation, betaherpesviruses are leading candidates for use as transmissible vaccine vectors. Here we evaluate the likely effectiveness of betaherpesvirus-vectored transmissible vaccines by developing and parameterizing a mathematical model using data from captive and free-living mouse populations infected with murine cytomegalovirus (MCMV). Simulations of our parameterized model demonstrate rapid and effective control for a range of pathogens, with pathogen elimination frequently occurring within a year of vaccine introduction. Our results also suggest, however, that the effectiveness of transmissible vaccines may vary across reservoir populations and with respect to the specific vector strain used to construct the vaccine.

Predicting the fine-scale spatial distribution of zoonotic reservoirs using computer vision.

Zoonotic diseases threaten human health worldwide and are often associated with anthropogenic disturbance. Predicting how disturbance influences spillover risk is critical for effective disease intervention but difficult to achieve at fine spatial scales. Here, we develop a method that learns the spatial distribution of a reservoir species from aerial imagery. Our approach uses neural networks to extract features of known or hypothesized importance from images. The spatial distribution of these features is then summarized and linked to spatially explicit reservoir presence/absence data using boosted regression trees. We demonstrate the utility of our method by applying it to the reservoir of Lassa virus, Mastomys natalensis, within the West African nations of Sierra Leone and Guinea. We show that, when trained using reservoir trapping data and publicly available aerial imagery, our framework learns relationships between environmental features and reservoir occurrence and accurately ranks areas according to the likelihood of reservoir presence.

Reservoir population ecology, viral evolution and the risk of emerging infectious disease.

The ecology and life history of wild animals influences their potential to harbour infectious disease. This observation has motivated studies identifying empirical relationships between traits of wild animals and historical patterns of spillover and emergence into humans. Although these studies have identified compelling broad-scale patterns, they are generally agnostic with respect to underlying mechanisms. Here, we develop mathematical models that couple reservoir population ecology with viral epidemiology and evolution to clarify existing verbal arguments and pinpoint the conditions that favour spillover and emergence. Our results support the idea that average lifespan influences the likelihood of an animal serving as a reservoir for human infectious disease. At the same time, however, our results show that the magnitude of this effect is sensitive to the rate of viral mutation. Our results also demonstrate that viral pathogens causing persistent infections or a transient immune response within the reservoir are more likely to fuel emergence. Genetically explicit stochastic simulations enrich these mathematical results by identifying relationships between the genetic basis of transmission and the risk of spillover and emergence. Together, our results clarify the scope of applicability for existing hypotheses and refine our understanding of emergence risk.

Bridging the gap: Using reservoir ecology and human serosurveys to estimate Lassa virus spillover in West Africa.

Forecasting the risk of pathogen spillover from reservoir populations of wild or domestic animals is essential for the effective deployment of interventions such as wildlife vaccination or culling. Due to the sporadic nature of spillover events and limited availability of data, developing and validating robust, spatially explicit, predictions is challenging. Recent efforts have begun to make progress in this direction by capitalizing on machine learning methodologies. An important weakness of existing approaches, however, is that they generally rely on combining human and reservoir infection data during the training process and thus conflate risk attributable to the prevalence of the pathogen in the reservoir population with the risk attributed to the realized rate of spillover into the human population. Because effective planning of interventions requires that these components of risk be disentangled, we developed a multi-layer machine learning framework that separates these processes. Our approach begins by training models to predict the geographic range of the primary reservoir and the subset of this range in which the pathogen occurs. The spillover risk predicted by the product of these reservoir specific models is then fit to data on realized patterns of historical spillover into the human population. The result is a geographically specific spillover risk forecast that can be easily decomposed and used to guide effective intervention. Applying our method to Lassa virus, a zoonotic pathogen that regularly spills over into the human population across West Africa, results in a model that explains a modest but statistically significant portion of geographic variation in historical patterns of spillover. When combined with a mechanistic mathematical model of infection dynamics, our spillover risk model predicts that 897,700 humans are infected by Lassa virus each year across West Africa, with Nigeria accounting for more than half of these human infections.

Microbial phenotypic heterogeneity in response to a metabolic toxin: Continuous, dynamically shifting distribution of formaldehyde tolerance in Methylobacterium extorquens populations.

While microbiologists often make the simplifying assumption that genotype determines phenotype in a given environment, it is becoming increasingly apparent that phenotypic heterogeneity (in which one genotype generates multiple phenotypes simultaneously even in a uniform environment) is common in many microbial populations. The importance of phenotypic heterogeneity has been demonstrated in a number of model systems involving binary phenotypic states (e.g., growth/non-growth); however, less is known about systems involving phenotype distributions that are continuous across an environmental gradient, and how those distributions change when the environment changes. Here, we describe a novel instance of phenotypic diversity in tolerance to a metabolic toxin within wild-type populations of Methylobacterium extorquens, a ubiquitous phyllosphere methylotroph capable of growing on the methanol periodically released from plant leaves. The first intermediate in methanol metabolism is formaldehyde, a potent cellular toxin that is lethal in high concentrations. We have found that at moderate concentrations, formaldehyde tolerance in M. extorquens is heterogeneous, with a cell's minimum tolerance level ranging between 0 mM and 8 mM. Tolerant cells have a distinct gene expression profile from non-tolerant cells. This form of heterogeneity is continuous in terms of threshold (the formaldehyde concentration where growth ceases), yet binary in outcome (at a given formaldehyde concentration, cells either grow normally or die, with no intermediate phenotype), and it is not associated with any detectable genetic mutations. Moreover, tolerance distributions within the population are dynamic, changing over time in response to growth conditions. We characterized this phenomenon using bulk liquid culture experiments, colony growth tracking, flow cytometry, single-cell time-lapse microscopy, transcriptomics, and genome resequencing. Finally, we used mathematical modeling to better understand the processes by which cells change phenotype, and found evidence for both stochastic, bidirectional phenotypic diversification and responsive, directed phenotypic shifts, depending on the growth substrate and the presence of toxin.

The role of alcohol consumption on acetaminophen induced liver injury: Implications from a mathematical model.

Acetaminophen (APAP) overdose is one of the predominant causes of drug induced acute liver injury in the U.S and U.K. Clinical studies show that ingestion of alcohol may increase the risk of APAP induced liver injury. Chronic alcoholism may potentiate APAP hepatotoxicity and this increased risk of APAP toxicity is observed when APAP is ingested even shortly after alcohol is cleared from the body. However, clinical reports also suggest that acute alcohol consumption may have a protective effect against hepatotoxicity by inhibiting microsomal acetaminophen oxidation and thereby reducing N-acetyl-p-benzoquinone imine (NAPQI) production. The aim of this study is to model this dual role of alcohol to determine how the timing of alcohol ingestion affects APAP metabolism and resulting liver injury and identify mechanisms of APAP induced liver injury. The mathematical model is developed to capture condition of a patient of single time APAP overdose who may be an acute or chronic alcohol user. The analysis suggests that the risk of APAP-induced hepatotoxicity is increased if APAP is ingested shortly after alcohol is cleared from the body in chronic alcohol users. A protective effect of acute consumption of alcohol is also observed in patients with APAP overdose. For example, simultaneous ingestion of alcohol and APAP overdose or alcohol intake after or before few hours of APAP overdose may result in less APAP-induced hepatotoxicity when compared to a single time APAP overdose. The rate of hepatocyte damage in APAP overdose patients depends on trade-off between induction and inhibition of CYP enzyme.

Woody cover and hominin environments in the past 6 million years.

The role of African savannahs in the evolution of early hominins has been debated for nearly a century. Resolution of this issue has been hindered by difficulty in quantifying the fraction of woody cover in the fossil record. Here we show that the fraction of woody cover in tropical ecosystems can be quantified using stable carbon isotopes in soils. Furthermore, we use fossil soils from hominin sites in the Awash and Omo-Turkana basins in eastern Africa to reconstruct the fraction of woody cover since the Late Miocene epoch (about 7 million years ago). (13)C/(12)C ratio data from 1,300 palaeosols at or adjacent to hominin sites dating to at least 6 million years ago show that woody cover was predominantly less than ∼40% at most sites. These data point to the prevalence of open environments at the majority of hominin fossil sites in eastern Africa over the past 6 million years.

Modeling the dynamics of stable isotope tissue-diet enrichment.

Reconstructions of dietary composition and trophic level from stable isotope measurements of animal tissue rely on predictable offsets of stable isotope ratios from diet to tissue. Physiological processes associated with metabolism shape tissue stable isotope ratios, and as such the spacing between stable isotope ratios of diet and tissue may be influenced by processes such as growth, nutritional stress, and disease. Here, we develop a model of incorporation stable isotopes in diet to tissues by coupling stable isotope dynamics to a model of macronutrient energy metabolism. We use the model to explore the effect of changes in dietary intake, both composition and amount, and in energy expenditure, on body mass and carbon and nitrogen stable isotope ratios of tissue.

Deconvolution of isotope signals from bundles of multiple hairs.

Segmental analysis of hair has been used in diverse fields ranging from forensics to ecology to measure the concentration of substances such as drugs and isotopes. Multiple hairs are typically combined into a bundle for segmental analysis to obtain a high-resolution series of measurements. Individual hair strands cycle through multiple phases of growth and grow at different rates when in the growth phase. Variation in growth of hair strands in a bundle can cause misalignment of substance concentration between hairs, attenuating the primary body signal. We developed a mathematical model based on the known physiology of hair growth to describe the signal averaging caused by bundling multiple hairs for segmental analysis. The model was used to form an inverse method to estimate the primary body signal from measurements of a hair bundle. The inverse method was applied to a previously described stable oxygen isotope chronology from the hair of a murder victim and provides a refined interpretation of the data. Aspects of the reconstruction were confirmed when the victim was later identified.

Recombinant transmissible vaccines will be intrinsically contained despite the ability to superinfect.

INTRODUCTION: Transmissible vaccines offer a novel approach to suppressing viruses in wildlife populations, with possible applications against viruses that infect humans as zoonoses - Lassa, Ebola, rabies. To ensure safety, current designs propose a recombinant vector platform in which the vector is isolated from the target wildlife population. Because using an endemic vector creates the potential for preexisting immunity to block vaccine transmission, these designs focus on vector viruses capable of superinfection, spreading throughout the host population following vaccination of few individuals. AREAS COVERED: We present original theoretical arguments that, regardless of its R0 value, a recombinant vaccine using a superinfecting vector is not expected to expand its active infection coverage when released into a wildlife population that already carries the vector. However, if superinfection occurs at a high rate such that individuals are repeatedly infected throughout their lives, the immunity footprint in the population can be high despite a low incidence of active vaccine infections. Yet we provide reasons that the above expectation is optimistic. EXPERT OPINION: High vaccine coverage will typically require repeated releases or release into a population lacking the vector, but careful attention to vector choice and vaccine engineering should also help improve transmissible vaccine utility.

Mathematical modeling of liver injury and dysfunction after acetaminophen overdose: early discrimination between survival and death.

UNLABELLED: Acetaminophen (APAP) is the leading cause of acute liver injury in the developed world. Timely administration of N-acetylcysteine (N-Ac) prevents the progression of serious liver injury and disease, whereas failure to administer N-Ac within a critical time frame allows disease progression and in the most severe cases may result in liver failure or death. In this situation, liver transplantation may be the only life-saving measure. Thus, the outcome of an APAP overdose depends on the size of the overdose and the time to first administration of N-Ac. We developed a system of differential equations to describe acute liver injury due to APAP overdose. The Model for Acetaminophen-induced Liver Damage (MALD) uses a patient's aspartate aminotransferase (AST), alanine aminotransferase (ALT), and international normalized ratio (INR) measurements on admission to estimate overdose amount, time elapsed since overdose, and outcome. The mathematical model was then tested on 53 patients from the University of Utah. With the addition of serum creatinine, eventual death was predicted with 100% sensitivity, 91% specificity, 67% positive predictive value (PPV), and 100% negative predictive value (NPV) in this retrospective study. Using only initial AST, ALT, and INR measurements, the model accurately predicted subsequent laboratory values for the majority of individual patients. This is the first dynamical rather than statistical approach to determine poor prognosis in patients with life-threatening liver disease due to APAP overdose. CONCLUSION: MALD provides a method to estimate overdose amount, time elapsed since overdose, and outcome from patient laboratory values commonly available on admission in cases of acute liver failure due to APAP overdose and should be validated in multicenter prospective evaluation.

Optimizing the delivery of self-disseminating vaccines in fluctuating wildlife populations.

Zoonotic pathogens spread by wildlife continue to spill into human populations and threaten human lives. A potential way to reduce this threat is by vaccinating wildlife species that harbor pathogens that are infectious to humans. Unfortunately, even in cases where vaccines can be distributed en masse as edible baits, achieving levels of vaccine coverage sufficient for pathogen elimination is rare. Developing vaccines that self-disseminate may help solve this problem by magnifying the impact of limited direct vaccination. Although models exist that quantify how well these self-disseminating vaccines will work when introduced into temporally stable wildlife populations, how well they will perform when introduced into populations with pronounced seasonal population dynamics remains unknown. Here we develop and analyze mathematical models of fluctuating wildlife populations that allow us to study how reservoir ecology, vaccine design, and vaccine delivery interact to influence vaccine coverage and opportunities for pathogen elimination. Our results demonstrate that the timing of vaccine delivery can make or break the success of vaccination programs. As a general rule, the effectiveness of self-disseminating vaccines is optimized by introducing after the peak of seasonal reproduction when the number of susceptible animals is near its maximum.

How Interactions During Viral-Viral Coinfection Can Shape Infection Kinetics.

Respiratory virus infections are a leading cause of disease worldwide with multiple viruses detected in 20-30% of cases and several viruses simultaneously circulating. Some infections with viral copathogens have been shown to result in reduced pathogenicity while other virus pairings can worsen disease. The mechanisms driving these dichotomous outcomes are likely variable and have only begun to be examined in the laboratory and clinic. To better understand viral-viral coinfections and predict potential mechanisms that result in distinct disease outcomes, we first systematically fit mathematical models to viral load data from ferrets infected with respiratory syncytial virus (RSV) followed by influenza A virus (IAV) after 3 days. The results suggested that IAV reduced the rate of RSV production while RSV reduced the rate of IAV infected cell clearance. We then explored the realm of possible dynamics for scenarios not examined experimentally, including different infection order, coinfection timing, interaction mechanisms, and viral pairings. IAV coinfection with rhinovirus (RV) or SARS-CoV-2 (CoV2) was examined by using human viral load data from single infections together with murine weight loss data from IAV-RV, RV-IAV, and IAV-CoV2 coinfections to guide the interpretation of the model results. Similar to the results with RSV-IAV coinfection, this analysis showed that the increased disease severity observed during murine IAV-RV or IAV-CoV2 coinfection was likely due to slower clearance of IAV infected cells by the other viruses. On the contrary, the improved outcome when IAV followed RV could be replicated when the rate of RV infected cell clearance was reduced by IAV. Simulating viral-viral coinfections in this way provides new insights about how viral-viral interactions can regulate disease severity during coinfection and yields testable hypotheses ripe for experimental evaluation.

How Interactions during Viral-Viral Coinfection Can Shape Infection Kinetics.

Respiratory viral infections are a leading global cause of disease with multiple viruses detected in 20-30% of cases, and several viruses simultaneously circulating. Some infections with unique viral copathogens result in reduced pathogenicity, while other viral pairings can worsen disease. The mechanisms driving these dichotomous outcomes are likely variable and have only begun to be examined in the laboratory and clinic. To better understand viral-viral coinfections and predict potential mechanisms that result in distinct disease outcomes, we first systematically fit mathematical models to viral load data from ferrets infected with respiratory syncytial virus (RSV), followed by influenza A virus (IAV) after 3 days. The results suggest that IAV reduced the rate of RSV production, while RSV reduced the rate of IAV infected cell clearance. We then explored the realm of possible dynamics for scenarios that had not been examined experimentally, including a different infection order, coinfection timing, interaction mechanisms, and viral pairings. IAV coinfection with rhinovirus (RV) or SARS-CoV-2 (CoV2) was examined by using human viral load data from single infections together with murine weight-loss data from IAV-RV, RV-IAV, and IAV-CoV2 coinfections to guide the interpretation of the model results. Similar to the results with RSV-IAV coinfection, this analysis shows that the increased disease severity observed during murine IAV-RV or IAV-CoV2 coinfection was likely due to the slower clearance of IAV-infected cells by the other viruses. The improved outcome when IAV followed RV, on the other hand, could be replicated when the rate of RV infected cell clearance was reduced by IAV. Simulating viral-viral coinfections in this way provides new insights about how viral-viral interactions can regulate disease severity during coinfection and yields testable hypotheses ripe for experimental evaluation.

Identifying the genetic basis of viral spillover using Lassa virus as a test case.

The rate at which zoonotic viruses spill over into the human population varies significantly over space and time. Remarkably, we do not yet know how much of this variation is attributable to genetic variation within viral populations. This gap in understanding arises because we lack methods of genetic analysis that can be easily applied to zoonotic viruses, where the number of available viral sequences is often limited, and opportunistic sampling introduces significant population stratification. Here, we explore the feasibility of using patterns of shared ancestry to correct for population stratification, enabling genome-wide association methods to identify genetic substitutions associated with spillover into the human population. Using a combination of phylogenetically structured simulations and Lassa virus sequences collected from humans and rodents in Sierra Leone, we demonstrate that existing methods do not fully correct for stratification, leading to elevated error rates. We also demonstrate, however, that the Type I error rate can be substantially reduced by confining the analysis to a less-stratified region of the phylogeny, even in an already-small dataset. Using this method, we detect two candidate single-nucleotide polymorphisms associated with spillover in the Lassa virus polymerase gene and provide generalized recommendations for the collection and analysis of zoonotic viruses.

Hydrogen and oxygen isotope ratios in body water and hair: modeling isotope dynamics in nonhuman primates.

The stable isotopic composition of drinking water, diet, and atmospheric oxygen influence the isotopic composition of body water ((2)H/(1)H, (18)O/(16)O expressed as δ(2) H and δ(18)O). In turn, body water influences the isotopic composition of organic matter in tissues, such as hair and teeth, which are often used to reconstruct historical dietary and movement patterns of animals and humans. Here, we used a nonhuman primate system (Macaca fascicularis) to test the robustness of two different mechanistic stable isotope models: a model to predict the δ(2)H and δ(18)O values of body water and a second model to predict the δ(2)H and δ(18)O values of hair. In contrast to previous human-based studies, use of nonhuman primates fed controlled diets allowed us to further constrain model parameter values and evaluate model predictions. Both models reliably predicted the δ(2)H and δ(18)O values of body water and of hair. Moreover, the isotope data allowed us to better quantify values for two critical variables in the models: the δ(2)H and δ(18)O values of gut water and the (18)O isotope fractionation associated with a carbonyl oxygen-water interaction in the gut (α(ow)). Our modeling efforts indicated that better predictions for body water and hair isotope values were achieved by making the isotopic composition of gut water approached that of body water. Additionally, the value of α(ow) was 1.0164, in close agreement with the only other previously measured observation (microbial spore cell walls), suggesting robustness of this fractionation factor across different biological systems.

History of Animals using Isotope Records (HAIR): a 6-year dietary history of one family of African elephants.

The dietary and movement history of individual animals can be studied using stable isotope records in animal tissues, providing insight into long-term ecological dynamics and a species niche. We provide a 6-year history of elephant diet by examining tail hair collected from 4 elephants in the same social family unit in northern Kenya. Sequential measurements of carbon, nitrogen, and hydrogen isotope rations in hair provide a weekly record of diet and water resources. Carbon isotope ratios were well correlated with satellite-based measurements of the normalized difference vegetation index (NDVI) of the region occupied by the elephants as recorded by the global positioning system (GPS) movement record; the absolute amount of C(4) grass consumption is well correlated with the maximum value of NDVI during individual wet seasons. Changes in hydrogen isotope ratios coincided very closely in time with seasonal fluctuations in rainfall and NDVI whereas diet shifts to relatively high proportions of grass lagged seasonal increases in NDVI by approximately 2 weeks. The peak probability of conception in the population occurred approximately 3 weeks after peak grazing. Spatial and temporal patterns of resource use show that the only period of pure browsing by the focal elephants was located in an over-grazed, communally managed region outside the protected area. The ability to extract time-specific longitudinal records on animal diets, and therefore the ecological history of an organism and its environment, provides an avenue for understanding the impact of climate dynamics and land-use change on animal foraging behavior and habitat relations.

Structural identifiability of the generalized Lotka-Volterra model for microbiome studies.

Population dynamic models can be used in conjunction with time series of species abundances to infer interactions. Understanding microbial interactions is a prerequisite for numerous goals in microbiome research, including predicting how populations change over time, determining how manipulations of microbiomes affect dynamics and designing synthetic microbiomes to perform tasks. As such, there is great interest in adapting population dynamic theory for microbial systems. Despite the appeal, numerous hurdles exist. One hurdle is that the data commonly obtained from DNA sequencing yield estimates of relative abundances, while population dynamic models such as the generalized Lotka-Volterra model track absolute abundances or densities. It is not clear whether relative abundance data alone can be used to infer parameters of population dynamic models such as the Lotka-Volterra model. We used structural identifiability analyses to determine the extent to which a time series of relative abundances can be used to parametrize the generalized Lotka-Volterra model. We found that only with absolute abundance data to accompany relative abundance estimates from sequencing can all parameters be uniquely identified. However, relative abundance data alone do contain information on relative interaction strengths, which is sufficient for many studies where the goal is to estimate key interactions and their effects on dynamics. Using synthetic data of a simple community for which we know the underlying structure, local practical identifiability analysis showed that modest amounts of both process and measurement error do not fundamentally affect these identifiability properties.