Spatio-temporal Bayesian network models with latent variables for revealing trophic dynamics and functional networks in fisheries ecology

The ADnet Bayesian belief network for alder decline: Integrating empirical data and expert knowledge

The globalization in plant material trading has caused the emergence of invasive pests in many ecosystems, such as the alder pathogen Phytophthora ×alni in European riparian forests. Due to the ecological importance of alder to the functioning of rivers and the increasing incidence of P. ×alni-induced alder decline, effective and accessible decision tools are required to help managers and stakeholders control the disease. This study proposes a Bayesian belief network methodology to integrate diverse information on the factors affecting the survival and infection ability of P. ×alni in riparian habitats to help predict and manage disease incidence. The resulting Alder Decline Network (ADnet) management tool integrates information about alder decline from scientific literature, expert knowledge and empirical data. Expert knowledge was gathered through elicitation techniques that included 19 experts from 12 institutions and 8 countries. An original dataset was created covering 1189 European locations, from which P. ×alni occurrence was modeled based on bioclimatic variables. ADnet uncertainty was evaluated through its sensitivity to changes in states and three scenario analyses. The ADnet tool indicated that mild temperatures and high precipitation are key factors favoring pathogen survival. Flood timing, water velocity, and soil type have the strongest influence on disease incidence. ADnet can support ecosystem management decisions and knowledge transfer to address P. ×alni-induced alder decline at local or regional levels across Europe. Management actions such as avoiding the planting of potentially infected trees or removing man-made structures that increase the flooding period in disease-affected sites could decrease the incidence of alder disease in riparian forests and limit its spread. The coverage of the ADnet tool can be expanded by updating data on the pathogen's occurrence, particularly from its distributional limits. Research on the role of genetic variability in alder susceptibility and pathogen virulence may also help improve future ADnet versions.

Untangling social-ecological interactions: A methods portfolio approach to tackling contemporary sustainability challenges in fisheries

Meeting the objectives of sustainable fisheries management requires attention to the complex interactions between humans, institutions and ecosystems that give rise to fishery outcomes. Traditional approaches to studying fisheries often do not fully capture, nor focus on these complex interactions between people and ecosystems. Despite advances in the scope and scale of interactions encompassed by more holistic methods, for example ecosystem-based fisheries management approaches, no single method can adequately capture the complexity of human-nature interactions. Approaches that combine quantitative and qualitative analytical approaches are necessary to generate a deeper understanding of these interactions and illuminate pathways to address fisheries sustainability challenges. However, combining methods is inherently challenging and requires understanding multiple methods from different, often disciplinarily distinct origins, demanding reflexivity of the researchers involved. Social-ecological systems' research has a history of utilising combinations of methods across the social and ecological realms to account for spatial and temporal dynamics, uncertainty and feedbacks that are key components of fisheries. We describe several categories of analytical methods (statistical modelling, network analysis, dynamic modelling, qualitative analysis and controlled behavioural experiments) and highlight their applications in fisheries research, strengths and limitations, data needs and overall objectives. We then discuss important considerations of a methods portfolio development process, including reflexivity, epistemological and ontological concerns and illustrate these considerations via three case studies. We show that, by expanding their methods portfolios, researchers will be better equipped to study the complex interactions shaping fisheries and contribute to solutions for sustainable fisheries management.

Optimising predictive models to prioritise viral discovery in zoonotic reservoirs

Despite the global investment in One Health disease surveillance, it remains difficult and costly to identify and monitor the wildlife reservoirs of novel zoonotic viruses. Statistical models can guide sampling target prioritisation, but the predictions from any given model might be highly uncertain; moreover, systematic model validation is rare, and the drivers of model performance are consequently under-documented. Here, we use the bat hosts of betacoronaviruses as a case study for the data-driven process of comparing and validating predictive models of probable reservoir hosts. In early 2020, we generated an ensemble of eight statistical models that predicted host-virus associations and developed priority sampling recommendations for potential bat reservoirs of betacoronaviruses and bridge hosts for SARS-CoV-2. During a time frame of more than a year, we tracked the discovery of 47 new bat hosts of betacoronaviruses, validated the initial predictions, and dynamically updated our analytical pipeline. We found that ecological trait-based models performed well at predicting these novel hosts, whereas network methods consistently performed approximately as well or worse than expected at random. These findings illustrate the importance of ensemble modelling as a buffer against mixed-model quality and highlight the value of including host ecology in predictive models. Our revised models showed an improved performance compared with the initial ensemble, and predicted more than 400 bat species globally that could be undetected betacoronavirus hosts. We show, through systematic validation, that machine learning models can help to optimise wildlife sampling for undiscovered viruses and illustrates how such approaches are best implemented through a dynamic process of prediction, data collection, validation, and updating.

Guidelines for model adaptation: A study of the transferability of a general seagrass ecosystem Dynamic Bayesian Networks model

In general, it is not feasible to collect enough empirical data to capture the entire range of processes that define a complex system, either intrinsically or when viewing the system from a different geographical or temporal perspective. In this context, an alternative approach is to consider model transferability, which is the act of translating a model built for one environment to another less well-known situation. Model transferability and adaptability may be extremely beneficial-approaches that aid in the reuse and adaption of models, particularly for sites with limited data, would benefit from widespread model uptake. Besides the reduced effort required to develop a model, data collection can be simplified when transferring a model to a different application context. The research presented in this paper focused on a case study to identify and implement guidelines for model adaptation. Our study adapted a general Dynamic Bayesian Networks (DBN) of a seagrass ecosystem to a new location where nodes were similar, but the conditional probability tables varied. We focused on two species of seagrass (Zostera noltei and Zostera marina) located in Arcachon Bay, France. Expert knowledge was used to complement peer-reviewed literature to identify which components needed adjustment including parameterization and quantification of the model and desired outcomes. We adopted both linguistic labels and scenario-based elicitation to elicit from experts the conditional probabilities used to quantify the DBN. Following the proposed guidelines, the model structure of the general DBN was retained, but the conditional probability tables were adapted for nodes that characterized the growth dynamics in Zostera spp. population located in Arcachon Bay, as well as the seasonal variation on their reproduction. Particular attention was paid to the light variable as it is a crucial driver of growth and physiology for seagrasses. Our guidelines provide a way to adapt a general DBN to specific ecosystems to maximize model reuse and minimize re-development effort. Especially important from a transferability perspective are guidelines for ecosystems with limited data, and how simulation and prior predictive approaches can be used in these contexts.

Food web assessments in the Baltic Sea: Models bridging the gap between indicators and policy needs

Ecosystem-based management requires understanding of food webs. Consequently, assessment of food web status is mandatory according to the European Union's Marine Strategy Framework Directive (MSFD) for EU Member States. However, how to best monitor and assess food webs in practise has proven a challenging question. Here, we review and assess the current status of food web indicators and food web models, and discuss whether the models can help addressing current shortcomings of indicator-based food web assessments, using the Baltic Sea as an example region. We show that although the MSFD food web assessment was designed to use food web indicators alone, they are currently poorly fit for the purpose, because they lack interconnectivity of trophic guilds. We then argue that the multiple food web models published for this region have a high potential to provide additional coherence to the definition of good environmental status, the evaluation of uncertainties, and estimates for unsampled indicator values, but we also identify current limitations that stand in the way of more formal implementation of this approach. We close with a discussion of which current models have the best capacity for this purpose in the Baltic Sea, and of the way forward towards the combination of measurable indicators and modelling approaches in food web assessments.

Identifying latent variables in Dynamic Bayesian Networks with bootstrapping applied to Type 2 Diabetes complication prediction

Predicting complications associated with complex disease is a challenging task given imbalanced and highly correlated disease complications along with unmeasured or latent factors. To analyse the complications associated with complex disease, this article attempts to deal with complex imbalanced clinical data, whilst determining the influence of latent variables within causal networks generated from the observation. This work proposes appropriate Intelligent Data Analysis methods for building Dynamic Bayesian networks with latent variables, applied to small-sized clinical data (a case of Type 2 Diabetes complications). First, it adopts a Time Series Bootstrapping approach to re-sample the rare complication class with a replacement with respect to the dynamics of disease progression. Then, a combination of the Induction Causation algorithm and Link Strength metric (which is called IC*LS approach) is applied on the bootstrapped data for incrementally identifying latent variables. The most highlighted contribution of this paper gained insight into the disease progression by interpreting the latent states (with respect to the associated distributions of complications). An exploration of inference methods along with confidence interval assessed the influences of these latent variables. The obtained results demonstrated an improvement in the prediction performance.

Predicting ecosystem components in the Gulf of Mexico and their responses to climate variability with a dynamic Bayesian network model

The Gulf of Mexico is an ecologically and economically important marine ecosystem that is affected by a variety of natural and anthropogenic pressures. These complex and interacting pressures, together with the dynamic environment of the Gulf, present challenges for the effective management of its resources. The recent adoption of Bayesian networks to ecology allows for the discovery and quantification of complex interactions from data after making only a few assumptions about observations of the system. In this study, we apply Bayesian network models, with different levels of structural complexity and a varying number of hidden variables to account for uncertainty when modeling ecosystem dynamics. From these models, we predict focal ecosystem components within the Gulf of Mexico. The predictive ability of the models varied with their structure. The model that performed best was parameterized through data-driven learning techniques and accounted for multiple ecosystem components’ associations and their interactions with human and natural pressures over time. Then, we altered sea surface temperature in the best performing model to explore the response of different ecosystem components to increased temperature. The magnitude and even direction of predicted responses varied by ecosystem components due to heterogeneity in driving factors and their spatial overlap. Our findings suggest that due to varying components’ sensitivity to drivers, changes in temperature will potentially lead to trade-offs in terms of population productivity. We were able to discover meaningful interactions between ecosystem components and their environment and show how sensitive these relationships are to climate perturbations, which increases our understanding of the potential future response of the system to increasing temperature. Our findings demonstrate that accounting for additional sources of variation, by incorporating multiple interactions and pressures in the model layout, has the potential for gaining deeper insights into the structure and dynamics of ecosystems.

Network-based metabolic analysis and microbial community modeling

Network inference is being applied to studies of microbial ecology to visualize and characterize microbial communities. Network representations can allow examination of the underlying organizational structure of a microbial community, and identification of key players or environmental conditions that influence community assembly and stability. Microbial co-association networks provide information on the dynamics of community structure as a function of time or other external variables. Community metabolic networks can provide a mechanistic link between species through identification of metabolite exchanges and species specific resource requirements. When used together, co-association networks and metabolic networks can provide a more in-depth view of the hidden rules that govern the stability and dynamics of microbial communities.