Unlocking ensemble ecosystem modelling for large and complex networks.

The potential effects of conservation actions on threatened species can be predicted using ensemble ecosystem models by forecasting populations with and without intervention. These model ensembles commonly assume stable coexistence of species in the absence of available data. However, existing ensemble-generation methods become computationally inefficient as the size of the ecosystem network increases, preventing larger networks from being studied. We present a novel sequential Monte Carlo sampling approach for ensemble generation that is orders of magnitude faster than existing approaches. We demonstrate that the methods produce equivalent parameter inferences, model predictions, and tightly constrained parameter combinations using a novel sensitivity analysis method. For one case study, we demonstrate a speed-up from 108 days to 6 hours, while maintaining equivalent ensembles. Additionally, we demonstrate how to identify the parameter combinations that strongly drive feasibility and stability, drawing ecological insight from the ensembles. Now, for the first time, larger and more realistic networks can be practically simulated and analysed.

LooplessFluxSampler: an efficient toolbox for sampling the loopless flux solution space of metabolic models.

BACKGROUND: Uniform random sampling of mass-balanced flux solutions offers an unbiased appraisal of the capabilities of metabolic networks. Unfortunately, it is impossible to avoid thermodynamically infeasible loops in flux samples when using convex samplers on large metabolic models. Current strategies for randomly sampling the non-convex loopless flux space display limited efficiency and lack theoretical guarantees. RESULTS: Here, we present LooplessFluxSampler, an efficient algorithm for exploring the loopless mass-balanced flux solution space of metabolic models, based on an Adaptive Directions Sampling on a Box (ADSB) algorithm. ADSB is rooted in the general Adaptive Direction Sampling (ADS) framework, specifically the Parallel ADS, for which theoretical convergence and irreducibility results are available for sampling from arbitrary distributions. By sampling directions that adapt to the target distribution, ADSB traverses more efficiently the sample space achieving faster mixing than other methods. Importantly, the presented algorithm is guaranteed to target the uniform distribution over convex regions, and it provably converges on the latter distribution over more general (non-convex) regions provided the sample can have full support. CONCLUSIONS: LooplessFluxSampler enables scalable statistical inference of the loopless mass-balanced solution space of large metabolic models. Grounded in a theoretically sound framework, this toolbox provides not only efficient but also reliable results for exploring the properties of the almost surely non-convex loopless flux space. Finally, LooplessFluxSampler includes a Markov Chain diagnostics suite for assessing the quality of the final sample and the performance of the algorithm.

Calibration of agent based models for monophasic and biphasic tumour growth using approximate Bayesian computation.

Agent-based models (ABMs) are readily used to capture the stochasticity in tumour evolution; however, these models are often challenging to validate with experimental measurements due to model complexity. The Voronoi cell-based model (VCBM) is an off-lattice agent-based model that captures individual cell shapes using a Voronoi tessellation and mimics the evolution of cancer cell proliferation and movement. Evidence suggests tumours can exhibit biphasic growth in vivo. To account for this phenomena, we extend the VCBM to capture the existence of two distinct growth phases. Prior work primarily focused on point estimation for the parameters without consideration of estimating uncertainty. In this paper, approximate Bayesian computation is employed to calibrate the model to in vivo measurements of breast, ovarian and pancreatic cancer. Our approach involves estimating the distribution of parameters that govern cancer cell proliferation and recovering outputs that match the experimental data. Our results show that the VCBM, and its biphasic extension, provides insight into tumour growth and quantifies uncertainty in the switching time between the two phases of the biphasic growth model. We find this approach enables precise estimates for the time taken for a daughter cell to become a mature cell. This allows us to propose future refinements to the model to improve accuracy, whilst also making conclusions about the differences in cancer cell characteristics.

Computationally efficient mechanism discovery for cell invasion with uncertainty quantification.

Parameter estimation for mathematical models of biological processes is often difficult and depends significantly on the quality and quantity of available data. We introduce an efficient framework using Gaussian processes to discover mechanisms underlying delay, migration, and proliferation in a cell invasion experiment. Gaussian processes are leveraged with bootstrapping to provide uncertainty quantification for the mechanisms that drive the invasion process. Our framework is efficient, parallelisable, and can be applied to other biological problems. We illustrate our methods using a canonical scratch assay experiment, demonstrating how simply we can explore different functional forms and develop and test hypotheses about underlying mechanisms, such as whether delay is present. All code and data to reproduce this work are available at https://github.com/DanielVandH/EquationLearning.jl.

Efficient inference and identifiability analysis for differential equation models with random parameters.

Heterogeneity is a dominant factor in the behaviour of many biological processes. Despite this, it is common for mathematical and statistical analyses to ignore biological heterogeneity as a source of variability in experimental data. Therefore, methods for exploring the identifiability of models that explicitly incorporate heterogeneity through variability in model parameters are relatively underdeveloped. We develop a new likelihood-based framework, based on moment matching, for inference and identifiability analysis of differential equation models that capture biological heterogeneity through parameters that vary according to probability distributions. As our novel method is based on an approximate likelihood function, it is highly flexible; we demonstrate identifiability analysis using both a frequentist approach based on profile likelihood, and a Bayesian approach based on Markov-chain Monte Carlo. Through three case studies, we demonstrate our method by providing a didactic guide to inference and identifiability analysis of hyperparameters that relate to the statistical moments of model parameters from independent observed data. Our approach has a computational cost comparable to analysis of models that neglect heterogeneity, a significant improvement over many existing alternatives. We demonstrate how analysis of random parameter models can aid better understanding of the sources of heterogeneity from biological data.

Being Bayesian in the 2020s: opportunities and challenges in the practice of modern applied Bayesian statistics.

Building on a strong foundation of philosophy, theory, methods and computation over the past three decades, Bayesian approaches are now an integral part of the toolkit for most statisticians and data scientists. Whether they are dedicated Bayesians or opportunistic users, applied professionals can now reap many of the benefits afforded by the Bayesian paradigm. In this paper, we touch on six modern opportunities and challenges in applied Bayesian statistics: intelligent data collection, new data sources, federated analysis, inference for implicit models, model transfer and purposeful software products. This article is part of the theme issue 'Bayesian inference: challenges, perspectives, and prospects'.

Many leveled ordinal models for frequency of alcohol and drug use.

Background: The numbers of days people consume alcohol and other drugs over a fixed interval, such as 28 days, are often collected in surveys of substance use. The presence of an upper bound on these variables can result in response distributions with "ceiling effects." Also, if some peoples' substance use behaviors are characterized by weekly patterns of use, summaries of substance days-of-use over longer periods can exhibit multiple modes.Objective: To highlight advantages of ordinal models with a separate level for each distinct survey response, for bounded, and potentially multimodal, count data.Methods: We fitted a Bayesian proportional odds ordinal model to longitudinal cannabis days-of-use reported by 443 individuals who used illicit drugs (206 female, 214 male, 23 non-binary). We specified an ordinal level for each unique response to allow the exact numeric distribution implied by the predicted ordinal response to be inferred. We then compared the fit of the proportional odds model with binomial, negative binomial, hurdle negative binomial and beta-binomial models.Results: Posterior predictive checks and the leave one out information criterion both suggested that the proportional odds model gave a better fit to the cannabis days-of-use data than the other models. Cannabis use among the target population declined during the COVID-19 pandemic in Australia, with the odds of a member of the population exceeding any specified frequency of cannabis use in Wave 4 estimated to be 73% lower than in Wave 1 (median odds ratio 0.27, 90% credible interval 0.19, 0.38).Conclusion: Ordinal models can be suitable for complex count data.

Bayesian prediction of winning times for elite swimming events.

To develop a statistical model of winning times for international swimming events with the aim of predicting winning time distributions and the probability of winning for the 2020 and 2024 Olympic Games. The data set included first and third place times from all individual swimming events from the Olympics and World Championships from 1990 to 2019. We compared different model formulations fitted with Bayesian inference to obtain predictive distributions; comparisons were based on mean percentage error in out-of-sample predictions of Olympics and World Championships winning swim times from 2011 to 2019. The Bayesian time series regression model, comprising auto-regressive and moving average terms and other predictors, had the smallest mean prediction error of 0.57% (CI 0.46-0.74%). For context, using the respective previous Olympics or World Championships winning time resulted in a mean prediction error of 0.70% (CI 0.59-0.82%). The Olympics were on average 0.5% (CI 0.3-0.7%) faster than World Championships over the study period. The model computes the posterior predictive distribution, which allows coaches and athletes to evaluate the probability of winning given an individual's swim time, and the probability of being faster or slower than the previous winning time or even the world record.

Estimating a novel stochastic model for within-field disease dynamics of banana bunchy top virus via approximate Bayesian computation.

The Banana Bunchy Top Virus (BBTV) is one of the most economically important vector-borne banana diseases throughout the Asia-Pacific Basin and presents a significant challenge to the agricultural sector. Current models of BBTV are largely deterministic, limited by an incomplete understanding of interactions in complex natural systems, and the appropriate identification of parameters. A stochastic network-based Susceptible-Infected-Susceptible model has been created which simulates the spread of BBTV across the subsections of a banana plantation, parameterising nodal recovery, neighbouring and distant infectivity across summer and winter. Findings from posterior results achieved through Markov Chain Monte Carlo approach to approximate Bayesian computation suggest seasonality in all parameters, which are influenced by correlated changes in inspection accuracy, temperatures and aphid activity. This paper demonstrates how the model may be used for monitoring and forecasting of various disease management strategies to support policy-level decision making.

Analysing the predictive capacity and dose-response of wellness in load monitoring.

This study aimed to identify the predictive capacity of wellness questionnaires on measures of training load using machine learning methods. The distributions of, and dose-response between, wellness and other load measures were also examined, offering insights into response patterns. Data (n= 14,109) were collated from an athlete management systems platform (Catapult Sports, Melbourne, Australia) and were split across three sports (cricket, rugby league and football) with data analysis conducted in R (Version 3.4.3). Wellness (sleep quality, readiness to train, general muscular soreness, fatigue, stress, mood, recovery rating and motivation) as the dependent variable, and sRPE, sRPE-TL and markers of external load (total distance and m.min-1) as independent variables were included for analysis. Classification and regression tree models showed high cross-validated error rates across all sports (i.e., > 0.89) and low model accuracy (i.e., < 5% of variance explained by each model) with similar results demonstrated using random forest models. These results suggest wellness items have limited predictive capacity in relation to internal and external load measures. This result was consistent despite varying statistical approaches (regression, classification and random forest models) and transformation of wellness scores. These findings indicate practitioners should exercise caution when interpreting and applying wellness responses.

Relative rate of change in cognitive score network dynamics via Bayesian hierarchical models reveal spatial patterns of neurodegeneration.

The degeneration of the human brain is a complex process, which often affects certain brain regions due to healthy aging or disease. This degeneration can be evaluated on regions of interest (ROI) in the brain through probabilistic networks and morphological estimates. Current approaches for finding such networks are limited to analyses at discrete neuropsychological stages, which cannot appropriately account for connectivity dynamics over the onset of cognitive deterioration, and morphological changes are seldom unified with connectivity networks, despite known dependencies. To overcome these limitations, a probabilistic wombling model is proposed to simultaneously estimate ROI cortical thickness and covariance networks contingent on rates of change in cognitive decline. This proposed model was applied to analyze longitudinal data from healthy control (HC) and Alzheimer's disease (AD) groups and found connection differences pertaining to regions, which play a crucial role in lasting cognitive impairment, such as the entorhinal area and temporal regions. Moreover, HC cortical thickness estimates were significantly higher than those in the AD group across all ROIs. The analyses presented in this work will help practitioners jointly analyze brain tissue atrophy at the ROI-level conditional on neuropsychological networks, which could potentially allow for more targeted therapeutic interventions.

Hindsight is 2020 vision: a characterisation of the global response to the COVID-19 pandemic.

BACKGROUND: The global impact of COVID-19 and the country-specific responses to the pandemic provide an unparalleled opportunity to learn about different patterns of the outbreak and interventions. We model the global pattern of reported COVID-19 cases during the primary response period, with the aim of learning from the past to prepare for the future. METHODS: Using Bayesian methods, we analyse the response to the COVID-19 outbreak for 158 countries for the period 22 January to 9 June 2020. This encompasses the period in which many countries imposed a variety of response measures and initial relaxation strategies. Instead of modelling specific intervention types and timings for each country explicitly, we adopt a stochastic epidemiological model including a feedback mechanism on virus transmission to capture complex nonlinear dynamics arising from continuous changes in community behaviour in response to rising case numbers. We analyse the overall effect of interventions and community responses across diverse regions. This approach mitigates explicit consideration of issues such as period of infectivity and public adherence to government restrictions. RESULTS: Countries with the largest cumulative case tallies are characterised by a delayed response, whereas countries that avoid substantial community transmission during the period of study responded quickly. Countries that recovered rapidly also have a higher case identification rate and small numbers of undocumented community transmission at the early stages of the outbreak. We also demonstrate that uncertainty in numbers of undocumented infections dramatically impacts the risk of multiple waves. Our approach is also effective at pre-empting potential flare-ups. CONCLUSIONS: We demonstrate the utility of modelling to interpret community behaviour in the early epidemic stages. Two lessons learnt that are important for the future are: i) countries that imposed strict containment measures early in the epidemic fared better with respect to numbers of reported cases; and ii) broader testing is required early in the epidemic to understand the magnitude of undocumented infections and recover rapidly. We conclude that clear patterns of containment are essential prior to relaxation of restrictions and show that modelling can provide insights to this end.

The effect of access to a designated interdisciplinary post-acute rehabilitation service on participant outcomes after brain injury.

OBJECTIVE: This study aimed to determine the influence of participation in a designated acquired brain injury (ABI) transitional rehabilitation service (ABI TRS) on outcome, in the context of a historical comparison group (HIST). Design: A cohort study, with retrospective comparison. Participants: 187 persons with ABI. Measures: The Depression, Anxiety and Stress Scale (DASS-21), Mayo-Portland Adaptability Index (MPAI-4) and Sydney Psychosocial and Reintegration Scale (SPRS) were completed at discharge and 3 months after discharge. Participation in the ABI TRS involved interdisciplinary rehabilitation, 2-4 times per week, for 3 months after hospital discharge. Results: There was evidence that at 3 months, participants with ABI TRS showed stabilized psychological wellbeing, and improvements in MPAI-4 ability and participation scores; in addition to improvements in SPRS occupational activity and living skills scores. Conclusion: A designated ABI TRS may improve the transition from hospital to home, and could form an important part of the brain injury rehabilitation continuum, between the inpatient and community setting.

Principles of Experimental Design for Big Data Analysis.

Big Datasets are endemic, but are often notoriously difficult to analyse because of their size, heterogeneity and quality. The purpose of this paper is to open a discourse on the potential for modern decision theoretic optimal experimental design methods, which by their very nature have traditionally been applied prospectively, to improve the analysis of Big Data through retrospective designed sampling in order to answer particular questions of interest. By appealing to a range of examples, it is suggested that this perspective on Big Data modelling and analysis has the potential for wide generality and advantageous inferential and computational properties. We highlight current hurdles and open research questions surrounding efficient computational optimisation in using retrospective designs, and in part this paper is a call to the optimisation and experimental design communities to work together in the field of Big Data analysis.

Melanoma Cell Colony Expansion Parameters Revealed by Approximate Bayesian Computation.

In vitro studies and mathematical models are now being widely used to study the underlying mechanisms driving the expansion of cell colonies. This can improve our understanding of cancer formation and progression. Although much progress has been made in terms of developing and analysing mathematical models, far less progress has been made in terms of understanding how to estimate model parameters using experimental in vitro image-based data. To address this issue, a new approximate Bayesian computation (ABC) algorithm is proposed to estimate key parameters governing the expansion of melanoma cell (MM127) colonies, including cell diffusivity, D, cell proliferation rate, λ, and cell-to-cell adhesion, q, in two experimental scenarios, namely with and without a chemical treatment to suppress cell proliferation. Even when little prior biological knowledge about the parameters is assumed, all parameters are precisely inferred with a small posterior coefficient of variation, approximately 2-12%. The ABC analyses reveal that the posterior distributions of D and q depend on the experimental elapsed time, whereas the posterior distribution of λ does not. The posterior mean values of D and q are in the ranges 226-268 µm2h-1, 311-351 µm2h-1 and 0.23-0.39, 0.32-0.61 for the experimental periods of 0-24 h and 24-48 h, respectively. Furthermore, we found that the posterior distribution of q also depends on the initial cell density, whereas the posterior distributions of D and λ do not. The ABC approach also enables information from the two experiments to be combined, resulting in greater precision for all estimates of D and λ.

Alive SMC(2) : Bayesian model selection for low-count time series models with intractable likelihoods.

In this article we present a new method for performing Bayesian parameter inference and model choice for low- count time series models with intractable likelihoods. The method involves incorporating an alive particle filter within a sequential Monte Carlo (SMC) algorithm to create a novel exact-approximate algorithm, which we refer to as alive SMC2. The advantages of this approach over competing methods are that it is naturally adaptive, it does not involve between-model proposals required in reversible jump Markov chain Monte Carlo, and does not rely on potentially rough approximations. The algorithm is demonstrated on Markov process and integer autoregressive moving average models applied to real biological datasets of hospital-acquired pathogen incidence, animal health time series, and the cumulative number of prion disease cases in mule deer.

Bayesian adaptive design: improving the effectiveness of monitoring of the Great Barrier Reef.

Monitoring programs are essential for understanding patterns, trends, and threats in ecological and environmental systems. However, such programs are costly in terms of dollars, human resources, and technology, and complex in terms of balancing short- and long-term requirements. In this work, We develop new statistical methods for implementing cost-effective adaptive sampling and monitoring schemes for coral reef that can better utilize existing information and resources, and which can incorporate available prior information. Our research was motivated by developing efficient monitoring practices for Australia's Great Barrier Reef. We develop and implement two types of adaptive sampling schemes, static and sequential, and show that they can be more informative and cost-effective than an existing (nonadaptive) monitoring program. Our methods are developed in a Bayesian framework with a range of utility functions relevant to environmental monitoring. Our results demonstrate the considerable potential for adaptive design to support improved management outcomes in comparison to set-and-forget styles of surveillance monitoring.

A new approach to estimating trends in chlamydia incidence.

OBJECTIVES: Directly measuring disease incidence in a population is difficult and not feasible to do routinely. We describe the development and application of a new method for estimating at a population level the number of incident genital chlamydia infections, and the corresponding incidence rates, by age and sex using routine surveillance data. METHODS: A Bayesian statistical approach was developed to calibrate the parameters of a decision-pathway tree against national data on numbers of notifications and tests conducted (2001-2013). Independent beta probability density functions were adopted for priors on the time-independent parameters; the shapes of these beta parameters were chosen to match prior estimates sourced from peer-reviewed literature or expert opinion. To best facilitate the calibration, multivariate Gaussian priors on (the logistic transforms of) the time-dependent parameters were adopted, using the Matérn covariance function to favour small changes over consecutive years and across adjacent age cohorts. The model outcomes were validated by comparing them with other independent empirical epidemiological measures, that is, prevalence and incidence as reported by other studies. RESULTS: Model-based estimates suggest that the total number of people acquiring chlamydia per year in Australia has increased by ∼120% over 12 years. Nationally, an estimated 356 000 people acquired chlamydia in 2013, which is 4.3 times the number of reported diagnoses. This corresponded to a chlamydia annual incidence estimate of 1.54% in 2013, increased from 0.81% in 2001 (∼90% increase). CONCLUSIONS: We developed a statistical method which uses routine surveillance (notifications and testing) data to produce estimates of the extent and trends in chlamydia incidence.

Model choice problems using approximate Bayesian computation with applications to pathogen transmission data sets.

Analytically or computationally intractable likelihood functions can arise in complex statistical inferential problems making them inaccessible to standard Bayesian inferential methods. Approximate Bayesian computation (ABC) methods address such inferential problems by replacing direct likelihood evaluations with repeated sampling from the model. ABC methods have been predominantly applied to parameter estimation problems and less to model choice problems due to the added difficulty of handling multiple model spaces. The ABC algorithm proposed here addresses model choice problems by extending Fearnhead and Prangle (2012, Journal of the Royal Statistical Society, Series B 74, 1-28) where the posterior mean of the model parameters estimated through regression formed the summary statistics used in the discrepancy measure. An additional stepwise multinomial logistic regression is performed on the model indicator variable in the regression step and the estimated model probabilities are incorporated into the set of summary statistics for model choice purposes. A reversible jump Markov chain Monte Carlo step is also included in the algorithm to increase model diversity for thorough exploration of the model space. This algorithm was applied to a validating example to demonstrate the robustness of the algorithm across a wide range of true model probabilities. Its subsequent use in three pathogen transmission examples of varying complexity illustrates the utility of the algorithm in inferring preference of particular transmission models for the pathogens.

Innovative approaches in soil carbon sequestration modelling for better prediction with limited data.

Soil carbon accounting and prediction play a key role in building decision support systems for land managers selling carbon credits, in the spirit of the Paris and Kyoto protocol agreements. Land managers typically rely on computationally complex models fit using sparse datasets to make these accounts and predictions. The model complexity and sparsity of the data can lead to over-fitting, leading to inaccurate results when making predictions with new data. Modellers address over-fitting by simplifying their models and reducing the number of parameters, and in the current context this could involve neglecting some soil organic carbon (SOC) components. In this study, we introduce two novel SOC models and a new RothC-like model and investigate how the SOC components and complexity of the SOC models affect the SOC prediction in the presence of small and sparse time series data. We develop model selection methods that can identify the soil carbon model with the best predictive performance, in light of the available data. Through this analysis we reveal that commonly used complex soil carbon models can over-fit in the presence of sparse time series data, and our simpler models can produce more accurate predictions.