Limits to Causal Inference with State-Space Reconstruction for Infectious Disease

Revealing causality in the associations between meteorological variables and air pollutant concentrations

Understanding the role of meteorology in determining air pollutant concentrations is an important goal for better comprehension of air pollution dispersion and fate. It requires estimating the strength of the causal associations between all the relevant meteorological variables and the pollutant concentrations. Unfortunately, many of the meteorological variables are not routinely observed. Furthermore, the common analysis methods cannot establish causality. Here we use the output of a numerical weather prediction model as a proxy for real meteorological data, and study the causal relationships between a large suite of its meteorological variables, including some rarely observed ones, and the corresponding nitrogen dioxide (NO2) concentrations at multiple observation locations. Time-lagged convergent cross mapping analysis is used to ascertain causality and its strength, and the Pearson and Spearman correlations are used to study the direction of the associations. The solar radiation, temperature lapse rate, boundary layer height, horizontal wind speed and wind shear were found to be causally associated with the NO2 concentrations, with mean time lags of their maximal impact at -3, -1, -2 and -3 hours, respectively. The nature of the association with the vertical wind speed was found to be uncertain and region-dependent. No causal association was found with relative humidity, temperature and precipitation.

Causal inference from cross-sectional earth system data with geographical convergent cross mapping

Causal inference in complex systems has been largely promoted by the proposal of some advanced temporal causation models. However, temporal models have serious limitations when time series data are not available or present insignificant variations, which causes a common challenge for earth system science. Meanwhile, there are few spatial causation models for fully exploring the rich spatial cross-sectional data in Earth systems. The generalized embedding theorem proves that observations can be combined together to construct the state space of the dynamic system, and if two variables are from the same dynamic system, they are causally linked. Inspired by this, here we show a Geographical Convergent Cross Mapping (GCCM) model for spatial causal inference with spatial cross-sectional data-based cross-mapping prediction in reconstructed state space. Three typical cases, where clearly existing causations cannot be measured through temporal models, demonstrate that GCCM could detect weak-moderate causations when the correlation is not significant. When the coupling between two variables is significant and strong, GCCM is advantageous in identifying the primary causation direction and better revealing the bidirectional asymmetric causation, overcoming the mirroring effect.

A general model-based causal inference method overcomes the curse of synchrony and indirect effect

To identify causation, model-free inference methods, such as Granger Causality, have been widely used due to their flexibility. However, they have difficulty distinguishing synchrony and indirect effects from direct causation, leading to false predictions. To overcome this, model-based inference methods that test the reproducibility of data with a specific mechanistic model to infer causality were developed. However, they can only be applied to systems described by a specific model, greatly limiting their applicability. Here, we address this limitation by deriving an easily testable condition for a general monotonic ODE model to reproduce time-series data. We built a user-friendly computational package, General ODE-Based Inference (GOBI), which is applicable to nearly any monotonic system with positive and negative regulations described by ODE. GOBI successfully inferred positive and negative regulations in various networks at both the molecular and population levels, unlike existing model-free methods. Thus, this accurate and broadly applicable inference method is a powerful tool for understanding complex dynamical systems.

Conditional cross-map-based technique: From pairwise dynamical causality to causal network reconstruction

Causality detection methods based on mutual cross mapping have been fruitfully developed and applied to data originating from nonlinear dynamical systems, where the causes and effects are non-separable. However, these pairwise methods still have shortcomings in discriminating typical network structures, including common drivers, indirect dependencies, and facing the curse of dimensionality, when they are stepping to causal network reconstruction. A few endeavors have been devoted to conquer these shortcomings. Here, we propose a novel method that could be regarded as one of these endeavors. Our method, named conditional cross-map-based technique, can eliminate third-party information and successfully detect direct dynamical causality, where the detection results can exactly be categorized into four standard normal forms by the designed criterion. To demonstrate the practical usefulness of our model-free, data-driven method, data generated from different representative models covering all kinds of network motifs and measured from real-world systems are investigated. Because correct identification of the direct causal links is essential to successful modeling, predicting, and controlling the underlying complex systems, our method does shed light on uncovering the inner working mechanisms of real-world systems only using the data experimentally obtained in a variety of disciplines.

Inferring a Causal Relationship between Environmental Factors and Respiratory Infections Using Convergent Cross-Mapping

The incidence of respiratory infections in the population is related to many factors, among which environmental factors such as air quality, temperature, and humidity have attracted much attention. In particular, air pollution has caused widespread discomfort and concern in developing countries. Although the correlation between respiratory infections and air pollution is well known, establishing causality between them remains elusive. In this study, by conducting theoretical analysis, we updated the procedure of performing the extended convergent cross-mapping (CCM, a method of causal inference) to infer the causality between periodic variables. Consistently, we validated this new procedure on the synthetic data generated by a mathematical model. For real data in Shaanxi province of China in the period of 1 January 2010 to 15 November 2016, we first confirmed that the refined method is applicable by investigating the periodicity of influenza-like illness cases, an air quality index, temperature, and humidity through wavelet analysis. We next illustrated that air quality (quantified by AQI), temperature, and humidity affect the daily influenza-like illness cases, and, in particular, the respiratory infection cases increased progressively with increased AQI with a time delay of 11 days.

Loss of Earth system resilience during early Eocene transient global warming events

Superimposed on long-term late Paleocene-early Eocene warming (~59 to 52 million years ago), Earth's climate experienced a series of abrupt perturbations, characterized by massive carbon input into the ocean-atmosphere system and global warming. Here, we examine the three most punctuated events of this period, the Paleocene-Eocene Thermal Maximum and Eocene Thermal Maximum 2 and 3, to probe whether they were initiated by climate-driven carbon cycle tipping points. Specifically, we analyze the dynamics of climate and carbon cycle indicators acquired from marine sediments to detect changes in Earth system resilience and to identify positive feedbacks. Our analyses suggest a loss of Earth system resilience toward all three events. Moreover, dynamic convergent cross mapping reveals intensifying coupling between the carbon cycle and climate during the long-term warming trend, supporting increasingly dominant climate forcing of carbon cycle dynamics during the Early Eocene Climatic Optimum when these recurrent global warming events became more frequent.

Loss of Earth system resilience during early Eocene transient global warming events

Superimposed on long-term late Paleocene-early Eocene warming (~59 to 52 million years ago), Earth's climate experienced a series of abrupt perturbations, characterized by massive carbon input into the ocean-atmosphere system and global warming. Here, we examine the three most punctuated events of this period, the Paleocene-Eocene Thermal Maximum and Eocene Thermal Maximum 2 and 3, to probe whether they were initiated by climate-driven carbon cycle tipping points. Specifically, we analyze the dynamics of climate and carbon cycle indicators acquired from marine sediments to detect changes in Earth system resilience and to identify positive feedbacks. Our analyses suggest a loss of Earth system resilience toward all three events. Moreover, dynamic convergent cross mapping reveals intensifying coupling between the carbon cycle and climate during the long-term warming trend, supporting increasingly dominant climate forcing of carbon cycle dynamics during the Early Eocene Climatic Optimum when these recurrent global warming events became more frequent.

Decomposing predictability to identify dominant causal drivers in complex ecosystems

Ecosystems are complex systems of various physical, biological, and chemical processes. Since ecosystem dynamics are composed of a mixture of different levels of stochasticity and nonlinearity, handling these data is a challenge for existing methods of time series-based causal inferences. Here, we show that, by harnessing contemporary machine learning approaches, the concept of Granger causality can be effectively extended to the analysis of complex ecosystem time series and bridge the gap between dynamical and statistical approaches. The central idea is to use an ensemble of fast and highly predictive artificial neural networks to select a minimal set of variables that maximizes the prediction of a given variable. It enables decomposition of the relationship among variables through quantifying the contribution of an individual variable to the overall predictive performance. We show how our approach, EcohNet, can improve interaction network inference for a mesocosm experiment and simulated ecosystems. The application of the method to a long-term lake monitoring dataset yielded interpretable results on the drivers causing cyanobacteria blooms, which is a serious threat to ecological integrity and ecosystem services. Since performance of EcohNet is enhanced by its predictive capabilities, it also provides an optimized forecasting of overall components in ecosystems. EcohNet could be used to analyze complex and hybrid multivariate time series in many scientific areas not limited to ecosystems.

Data-driven causal analysis of observational biological time series

Complex systems are challenging to understand, especially when they defy manipulative experiments for practical or ethical reasons. Several fields have developed parallel approaches to infer causal relations from observational time series. Yet, these methods are easy to misunderstand and often controversial. Here, we provide an accessible and critical review of three statistical causal discovery approaches (pairwise correlation, Granger causality, and state space reconstruction), using examples inspired by ecological processes. For each approach, we ask what it tests for, what causal statement it might imply, and when it could lead us astray. We devise new ways of visualizing key concepts, describe some novel pathologies of existing methods, and point out how so-called ‘model-free’ causality tests are not assumption-free. We hope that our synthesis will facilitate thoughtful application of methods, promote communication across different fields, and encourage explicit statements of assumptions. A video walkthrough is available (Video 1 or https://youtu.be/AlV0ttQrjK8).

Inferring causal relations from observational long-term carbon and water fluxes records

Land, atmosphere and climate interact constantly and at different spatial and temporal scales. In this paper we rely on causal discovery methods to infer spatial patterns of causal relations between several key variables of the carbon and water cycles: gross primary productivity, latent heat energy flux for evaporation, surface air temperature, precipitation, soil moisture and radiation. We introduce a methodology based on the convergent cross-mapping (CCM) technique. Despite its good performance in general, CCM is sensitive to (even moderate) noise levels and hyper-parameter selection. We present a robust CCM (RCCM) that relies on temporal bootstrapping decision scores and the derivation of more stringent cross-map skill scores. The RCCM method is combined with the information-geometric causal inference (IGCI) method to address the problem of strong and instantaneous variable coupling, another important and long-standing issue of CCM. The proposed methodology allows to derive spatially explicit global maps of causal relations between the involved variables and retrieve the underlying complexity of the interactions. Results are generally consistent with reported patterns and process understanding, and constitute a new way to quantify and understand carbon and water fluxes interactions.

Inferring causality in biological oscillators

MOTIVATION

Fundamental to biological study is identifying regulatory interactions. The recent surge in time-series data collection in biology provides a unique opportunity to infer regulations computationally. However, when components oscillate, model-free inference methods, while easily implemented, struggle to distinguish periodic synchrony and causality. Alternatively, model-based methods test the reproducibility of time series given a specific model but require inefficient simulations and have limited applicability.

RESULTS

We develop an inference method based on a general model of molecular, neuronal and ecological oscillatory systems that merges the advantages of both model-based and model-free methods, namely accuracy, broad applicability and usability. Our method successfully infers the positive and negative regulations within various oscillatory networks, e.g. the repressilator and a network of cofactors at the pS2 promoter, outperforming popular inference methods.

AVAILABILITY AND IMPLEMENTATION

We provide a computational package, ION (Inferring Oscillatory Networks), that users can easily apply to noisy, oscillatory time series to uncover the mechanisms by which diverse systems generate oscillations. Accompanying MATLAB code under a BSD-style license and examples are available at https://github.com/Mathbiomed/ION. Additionally, the code is available under a CC-BY 4.0 License at https://doi.org/10.6084/m9.figshare.16431408.v1.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Susceptible host availability modulates climate effects on dengue dynamics

Experiments and models suggest that climate affects mosquito-borne disease transmission. However, disease transmission involves complex nonlinear interactions between climate and population dynamics, which makes detecting climate drivers at the population level challenging. By analysing incidence data, estimated susceptible population size, and climate data with methods based on nonlinear time series analysis (collectively referred to as empirical dynamic modelling), we identified drivers and their interactive effects on dengue dynamics in San Juan, Puerto Rico. Climatic forcing arose only when susceptible availability was high: temperature and rainfall had net positive and negative effects respectively. By capturing mechanistic, nonlinear and context-dependent effects of population susceptibility, temperature and rainfall on dengue transmission empirically, our model improves forecast skill over recent, state-of-the-art models for dengue incidence. Together, these results provide empirical evidence that the interdependence of host population susceptibility and climate drives dengue dynamics in a nonlinear and complex, yet predictable way.

An Empirical Test of the Role of Small-Scale Transmission in Large-Scale Disease Dynamics

A key assumption of epidemiological models is that population-scale disease spread is driven by close contact between hosts and pathogens. At larger scales, however, mechanisms such as spatial structure in host and pathogen populations and environmental heterogeneity could alter disease spread. The assumption that small-scale transmission mechanisms are sufficient to explain large-scale infection rates, however, is rarely tested. Here, we provide a rigorous test using an insect-baculovirus system. We fit a mathematical model to data from forest-wide epizootics while constraining the model parameters with data from branch-scale experiments, a difference in spatial scale of four orders of magnitude. This experimentally constrained model fits the epizootic data well, supporting the role of small-scale transmission, but variability is high. We then compare this model's performance to an unconstrained model that ignores the experimental data, which serves as a proxy for models with additional mechanisms. The unconstrained model has a superior fit, revealing a higher transmission rate across forests compared with branch-scale estimates. Our study suggests that small-scale transmission is insufficient to explain baculovirus epizootics. Further research is needed to identify the mechanisms that contribute to disease spread across large spatial scales, and synthesizing models and multiscale data are key to understanding these dynamics.

Stochasticity and Infectious Disease Dynamics: Density and Weather Effects on a Fungal Insect Pathogen

In deterministic models of epidemics, there is a host abundance threshold above which the introduction of a few infected individuals leads to a severe epidemic. Studies of weather-driven animal pathogens often assume that abundance thresholds will be overwhelmed by weather-driven stochasticity, but tests of this assumption are lacking. We collected observational and experimental data for a fungal pathogen, Entomophaga maimaiga, that infects the gypsy moth, Lymantria dispar. We used an advanced statistical-computing algorithm to fit mechanistic models to our data, such that different models made different assumptions about the effects of host density and weather on E. maimaiga epizootics (epidemics in animals). We then used Akaike information criterion analysis to choose the best model. In the best model, epizootics are driven by a combination of weather and host density, and the model does an excellent job of explaining the data, whereas models that allow only for weather effects or only for density-dependent effects do a poor job of explaining the data. Density-dependent transmission in our best model produces a host density threshold, but this threshold is strongly blurred by the stochastic effects of weather. Our work shows that host-abundance thresholds may be important even if weather strongly affects transmission, suggesting that epidemiological models that allow for weather have an important role to play in understanding animal pathogens. The success of our model means that it could be useful for managing the gypsy moth, an important pest of hardwood forests in North America.

Combined Effects of Natural Enemies and Competition for Resources on a Forest Defoliator: A Theoretical and Empirical Analysis

Explanations for the dynamics of insect outbreaks often focus on natural enemies, on the grounds that parasitoid and pathogen attack rates are high during outbreaks. While natural enemy models can successfully reproduce outbreak cycles, experiments have repeatedly demonstrated the importance of resource quality and abundance. Experiments, however, are rarely invoked in modeling studies. Here we combine mechanistic models, observational data, and field experiments to quantify the roles of parasitoid attacks and resource competition on the jack pine budworm, Choristoneura pinus. By fitting models to a combination of observational and experimental data, we show that parasitoid attacks are the main source of larval budworm mortality at low and intermediate budworm densities but that resource competition is the main source of mortality at high densities. Our results further show that the effects of resource competition become more severe with increasing host tree age and that the effects of parasitoids are moderated by strong competition between parasitoids for hosts. Allowing for these effects in a model of insect outbreaks leads to realistic outbreak cycles, while a host-parasitoid model without resource competition produces an unrealistic stable equilibrium. The effects of resource competition are modulated by tree age, which in turn depends on fire regimes. Our model therefore suggests that increases in fire frequency due to climate change may interact in complex ways with budworm outbreaks. Our work shows that resource competition can be as important as natural enemies in modulating insect outbreaks, while demonstrating the usefulness of high-performance computing in experimental field ecology.

Can Community Structure Causally Determine Dynamics of Constituent Species? A Test Using a Host-Parasite Community

Structures of communities have been widely studied with the assumption that they not only are a useful bookkeeping tool but also can causally influence dynamics of the populations from which they emerge. However, convincing tests of this assumption have remained elusive because generally the only way to alter a community property is by manipulating its constituent populations, thereby preventing independent measurements of effects on those populations. There is a growing body of evidence that methods like convergent cross-mapping (CCM) can be used to make inferences about causal interactions using state space reconstructions of coupled time series, a method that relies on only observational data. Here we show that CCM can be used to test the causal effects of community properties using a well-studied Slovakian rodent-ectoparasite community. CCM identified causal drivers across the organizational scales of this community, including evidence that host dynamics were influenced by the degree to which the community at large was connected and clustered. Our findings add to the growing literature on the importance of community structures in disease dynamics and argue for a broader use of causal inference in the analysis of community dynamics.

Climate drivers of malaria at its southern fringe in the Americas

In this work we analyze potential environmental drivers of malaria cases in Northwestern Argentina. We inspect causal links between malaria and climatic variables by means of the convergent cross mapping technique, which provides a causality criterion from the theory of dynamic systems. Analysis is based on 12 years of weekly malaria P. vivax cases in Tartagal, Salta, Argentina-at the southern fringe of malaria incidence in the Americas-together with humidity and temperature time-series spanning the same period. Our results show that there are causal links between malaria cases and both maximum temperature, with a delay of five weeks, and minimum temperature, with delays of zero and twenty two weeks. Humidity is also a driver of malaria cases, with thirteen weeks delay between cause and effect. Furthermore we also determined the sign and strength of the effects. Temperature has always a positive non-linear effect on cases, with maximum temperature effects more pronounced above 25°C and minimum above 17°C, while effects of humidity are more intricate: maximum humidity above 85% has a negative effect, whereas minimum humidity has a positive effect on cases. These results might be signaling processes operating at short (below 5 weeks) and long (over 12 weeks) time delays, corresponding to effects related to parasite cycle and mosquito population dynamics respectively. The non-linearities found for the strength of the effect of temperature on malaria cases make warmer areas more prone to higher increases in the disease incidence. Moreover, our results indicate that an increase of extreme weather events could enhance the risks of malaria spreading and re-emergence beyond the current distribution. Both situations, warmer climate and increase of extreme events, will be remarkably increased by the end of the century in this hot spot of climate change.

Comparison of six methods for the detection of causality in a bivariate time series

In this comparative study, six causality detection methods were compared, namely, the Granger vector autoregressive test, the extended Granger test, the kernel version of the Granger test, the conditional mutual information (transfer entropy), the evaluation of cross mappings between state spaces, and an assessment of predictability improvement due to the use of mixed predictions. Seven test data sets were analyzed: linear coupling of autoregressive models, a unidirectional connection of two Hénon systems, a unidirectional connection of chaotic systems of Rössler and Lorenz type and of two different Rössler systems, an example of bidirectionally connected two-species systems, a fishery model as an example of two correlated observables without a causal relationship, and an example of mediated causality. We tested not only 20000 points long clean time series but also noisy and short variants of the data. The standard and the extended Granger tests worked only for the autoregressive models. The remaining methods were more successful with the more complex test examples, although they differed considerably in their capability to reveal the presence and the direction of coupling and to distinguish causality from mere correlation.

Identifying climate drivers of infectious disease dynamics: recent advances and challenges ahead

Climate change is likely to profoundly modulate the burden of infectious diseases. However, attributing health impacts to a changing climate requires being able to associate changes in infectious disease incidence with the potentially complex influences of climate. This aim is further complicated by nonlinear feedbacks inherent in the dynamics of many infections, driven by the processes of immunity and transmission. Here, we detail the mechanisms by which climate drivers can shape infectious disease incidence, from direct effects on vector life history to indirect effects on human susceptibility, and detail the scope of variation available with which to probe these mechanisms. We review approaches used to evaluate and quantify associations between climate and infectious disease incidence, discuss the array of data available to tackle this question, and detail remaining challenges in understanding the implications of climate change for infectious disease incidence. We point to areas where synthesis between approaches used in climate science and infectious disease biology provide potential for progress.

Ecological Network Inference From Long-Term Presence-Absence Data

Ecological communities are characterized by complex networks of trophic and nontrophic interactions, which shape the dy-namics of the community. Machine learning and correlational methods are increasingly popular for inferring networks from co-occurrence and time series data, particularly in microbial systems. In this study, we test the suitability of these methods for inferring ecological interactions by constructing networks using Dynamic Bayesian Networks, Lasso regression, and Pear-son’s correlation coefficient, then comparing the model networks to empirical trophic and nontrophic webs in two ecological systems. We find that although each model significantly replicates the structure of at least one empirical network, no model significantly predicts network structure in both systems, and no model is clearly superior to the others. We also find that networks inferred for the Tatoosh intertidal match the nontrophic network much more closely than the trophic one, possibly due to the challenges of identifying trophic interactions from presence-absence data. Our findings suggest that although these methods hold some promise for ecological network inference, presence-absence data does not provide enough signal for models to consistently identify interactions, and networks inferred from these data should be interpreted with caution.