Ethical frameworks should be applied to computational modelling of infectious disease interventions.

This perspective is part of an international effort to improve epidemiological models with the goal of reducing the unintended consequences of infectious disease interventions. The scenarios in which models are applied often involve difficult trade-offs that are well recognised in public health ethics. Unless these trade-offs are explicitly accounted for, models risk overlooking contested ethical choices and values, leading to an increased risk of unintended consequences. We argue that such risks could be reduced if modellers were more aware of ethical frameworks and had the capacity to explicitly account for the relevant values in their models. We propose that public health ethics can provide a conceptual foundation for developing this capacity. After reviewing relevant concepts in public health and clinical ethics, we discuss examples from the COVID-19 pandemic to illustrate the current separation between public health ethics and infectious disease modelling. We conclude by describing practical steps to build the capacity for ethically aware modelling. Developing this capacity constitutes a critical step towards ethical practice in computational modelling of public health interventions, which will require collaboration with experts on public health ethics, decision support, behavioural interventions, and social determinants of health, as well as direct consultation with communities and policy makers.

Rapid prototyping of models for COVID-19 outbreak detection in workplaces.

Early case detection is critical to preventing onward transmission of COVID-19 by enabling prompt isolation of index infections, and identification and quarantining of contacts. Timeliness and completeness of ascertainment depend on the surveillance strategy employed. This paper presents modelling used to inform workplace testing strategies for the Australian government in early 2021. We use rapid prototype modelling to quickly investigate the effectiveness of testing strategies to aid decision making. Models are developed with a focus on providing relevant results to policy makers, and these models are continually updated and improved as new questions are posed. Developed to support the implementation of testing strategies in high risk workplace settings in Australia, our modelling explores the effects of test frequency and sensitivity on outbreak detection. We start with an exponential growth model, which demonstrates how outbreak detection changes depending on growth rate, test frequency and sensitivity. From the exponential model, we learn that low sensitivity tests can produce high probabilities of detection when testing occurs frequently. We then develop a more complex Agent Based Model, which was used to test the robustness of the results from the exponential model, and extend it to include intermittent workplace scheduling. These models help our fundamental understanding of disease detectability through routine surveillance in workplaces and evaluate the impact of testing strategies and workplace characteristics on the effectiveness of surveillance. This analysis highlights the risks of particular work patterns while also identifying key testing strategies to best improve outbreak detection in high risk workplaces.

Individual variation in vaccine immune response can produce bimodal distributions of protection.

The ability for vaccines to protect against infectious diseases varies among individuals, but computational models employed to inform policy typically do not account for this variation. Here we examine this issue: we implement a model of vaccine efficacy developed in the context of SARS-CoV-2 in order to evaluate the general implications of modelling correlates of protection on the individual level. Due to high levels of variation in immune response, the distributions of individual-level protection emerging from this model tend to be highly dispersed, and are often bimodal. We describe the specification of the model, provide an intuitive parameterisation, and comment on its general robustness. We show that the model can be viewed as an intermediate between the typical approaches that consider the mode of vaccine action to be either "all-or-nothing" or "leaky". Our view based on this analysis is that individual variation in correlates of protection is an important consideration that may be crucial to designing and implementing models for estimating population-level impacts of vaccination programs.

Modelling mass drug administration strategies for reducing scabies burden in Monrovia, Liberia.

Scabies is a parasitic infestation with high global burden. Mass drug administrations (MDAs) are recommended for communities with a scabies prevalence of >10%. Quantitative analyses are needed to demonstrate the likely effectiveness of MDA recommendations. In this study, we developed an agent-based model of scabies transmission calibrated to demographic and epidemiological data from Monrovia. We used this model to compare the effectiveness of MDA scenarios for achieving scabies elimination and reducing scabies burden, as measured by time until recrudescence following delivery of an MDA and disability-adjusted-life-years (DALYs) averted. Our model showed that three rounds of MDA delivered at six-month intervals and reaching 80% of the population could reduce prevalence below 2% for three years following the final round, before recrudescence. When MDAs were followed by increased treatment uptake, prevalence was maintained below 2% indefinitely. Increasing the number of and coverage of MDA rounds increased the probability of achieving elimination and the number of DALYs averted. Our results suggest that acute reduction of scabies prevalence by MDA can support a transition to improved treatment access. This study demonstrates how modelling can be used to estimate the expected impact of MDAs by projecting future epidemiological dynamics and health gains under alternative scenarios.

COVID-19 vaccine coverage targets to inform reopening plans in a low incidence setting.

Since the emergence of SARS-CoV-2 in 2019 through to mid-2021, much of the Australian population lived in a COVID-19-free environment. This followed the broadly successful implementation of a strong suppression strategy, including international border closures. With the availability of COVID-19 vaccines in early 2021, the national government sought to transition from a state of minimal incidence and strong suppression activities to one of high vaccine coverage and reduced restrictions but with still-manageable transmission. This transition is articulated in the national 're-opening' plan released in July 2021. Here, we report on the dynamic modelling study that directly informed policies within the national re-opening plan including the identification of priority age groups for vaccination, target vaccine coverage thresholds and the anticipated requirements for continued public health measures-assuming circulation of the Delta SARS-CoV-2 variant. Our findings demonstrated that adult vaccine coverage needed to be at least 60% to minimize public health and clinical impacts following the establishment of community transmission. They also supported the need for continued application of test-trace-isolate-quarantine and social measures during the vaccine roll-out phase and beyond.

Anomaly Detection in Endemic Disease Surveillance Data Using Machine Learning Techniques.

Disease surveillance is used to monitor ongoing control activities, detect early outbreaks, and inform intervention priorities and policies. However, data from disease surveillance that could be used to support real-time decisionmaking remain largely underutilised. Using the Brazilian Amazon malaria surveillance dataset as a case study, in this paper we explore the potential for unsupervised anomaly detection machine learning techniques to discover signals of epidemiological interest. We found that our models were able to provide an early indication of outbreak onset, outbreak peaks, and change points in the proportion of positive malaria cases. Specifically, the sustained rise in malaria in the Brazilian Amazon in 2016 was flagged by several models. We found that no single model detected all anomalies across all health regions. Because of this, we provide the minimum number of machine learning models top-k models) to maximise the number of anomalies detected across different health regions. We discovered that the top three models that maximise the coverage of the number and types of anomalies detected across the thirteen health regions are principal component analysis, stochastic outlier selection, and the minimum covariance determinant. Anomaly detection is a potentially valuable approach to discovering patterns of epidemiological importance when confronted with a large volume of data across space and time. Our exploratory approach can be replicated for other diseases and locations to inform monitoring, timely interventions, and actions towards the goal of controlling endemic disease.

Correlation of viral loads in disease transmission could affect early estimates of the reproduction number.

Early estimates of the transmission properties of a newly emerged pathogen are critical to an effective public health response, and are often based on limited outbreak data. Here, we use simulations to investigate how correlations between the viral load of cases in transmission chains can affect estimates of these fundamental transmission properties. Our computational model simulates a disease transmission mechanism in which the viral load of the infector at the time of transmission influences the infectiousness of the infectee. These correlations in transmission pairs produce a population-level convergence process during which the distributions of initial viral loads in each subsequent generation converge to a steady state. We find that outbreaks arising from index cases with low initial viral loads give rise to early estimates of transmission properties that could be misleading. These findings demonstrate the potential for transmission mechanisms to affect estimates of the transmission properties of newly emerged viruses in ways that could be operationally significant to a public health response.

Value of information analysis for pandemic response: intensive care unit preparedness at the onset of COVID-19.

BACKGROUND: During the early stages of the COVID-19 pandemic, there was considerable uncertainty surrounding epidemiological and clinical aspects of SARS-CoV-2. Governments around the world, starting from varying levels of pandemic preparedness, needed to make decisions about how to respond to SARS-CoV-2 with only limited information about transmission rates, disease severity and the likely effectiveness of public health interventions. In the face of such uncertainties, formal approaches to quantifying the value of information can help decision makers to prioritise research efforts. METHODS: In this study we use Value of Information (VoI) analysis to quantify the likely benefit associated with reducing three key uncertainties present in the early stages of the COVID-19 pandemic: the basic reproduction number ([Formula: see text]), case severity (CS), and the relative infectiousness of children compared to adults (CI). The specific decision problem we consider is the optimal level of investment in intensive care unit (ICU) beds. Our analysis incorporates mathematical models of disease transmission and clinical pathways in order to estimate ICU demand and disease outcomes across a range of scenarios. RESULTS: We found that VoI analysis enabled us to estimate the relative benefit of resolving different uncertainties about epidemiological and clinical aspects of SARS-CoV-2. Given the initial beliefs of an expert, obtaining more information about case severity had the highest parameter value of information, followed by the basic reproduction number [Formula: see text]. Resolving uncertainty about the relative infectiousness of children did not affect the decision about the number of ICU beds to be purchased for any COVID-19 outbreak scenarios defined by these three parameters. CONCLUSION: For the scenarios where the value of information was high enough to justify monitoring, if CS and [Formula: see text] are known, management actions will not change when we learn about child infectiousness. VoI is an important tool for understanding the importance of each disease factor during outbreak preparedness and can help to prioritise the allocation of resources for relevant information.

Evaluating the role of asymptomatic throat carriage of Streptococcus pyogenes in impetigo transmission in remote Aboriginal communities in Northern Territory, Australia: a retrospective genomic analysis.

BACKGROUND: Streptococcus pyogenes, or group A Streptococcus (GAS), infections contribute to a high burden of disease in Aboriginal Australians, causing skin infections and immune sequelae such as rheumatic heart disease. Controlling skin infections in these populations has proven difficult, with transmission dynamics being poorly understood. We aimed to identify the relative contributions of impetigo and asymptomatic throat carriage to GAS transmission. METHODS: In this genomic analysis, we retrospectively applied whole genome sequencing to GAS isolates that were collected as part of an impetigo surveillance longitudinal household survey conducted in three remote Aboriginal communities in the Northern Territory of Australia between Aug 6, 2003, and June 22, 2005. We included GAS isolates from all throats and impetigo lesions of people living in two of the previously studied communities. We classified isolates into genomic lineages based on pairwise shared core genomes of more than 99% with five or fewer single nucleotide polymorphisms. We used a household network analysis of epidemiologically and genomically linked lineages to quantify the transmission of GAS within and between households. FINDINGS: We included 320 GAS isolates in our analysis: 203 (63%) from asymptomatic throat swabs and 117 (37%) from impetigo lesions. Among 64 genomic lineages (encompassing 39 emm types) we identified 264 transmission links (involving 93% of isolates), for which the probable source was asymptomatic throat carriage in 166 (63%) and impetigo lesions in 98 (37%). Links originating from impetigo cases were more frequent between households than within households. Households were infected with GAS for a mean of 57 days (SD 39 days), and once cleared, reinfected 62 days (SD 40 days) later. Increased household size and community presence of GAS and scabies were associated with slower clearance of GAS. INTERPRETATION: In communities with high prevalence of endemic GAS-associated skin infection, asymptomatic throat carriage is a GAS reservoir. Public health interventions such as vaccination or community infection control programmes aimed at interrupting transmission of GAS might need to include consideration of asymptomatic throat carriage. FUNDING: Australian National Health and Medical Research Council.

Model-based estimation of the impact on rotavirus disease of RV3-BB vaccine administered in a neonatal or infant schedule.

Rotavirus infection is a common cause of severe diarrheal disease and a major cause of deaths and hospitalizations among young children. Incidence of rotavirus has declined globally with increasing vaccine coverage. However, it remains a significant cause of morbidity and mortality in low-income countries where vaccine access is limited and efficacy is lower. The oral human neonatal vaccine RV3-BB can be safely administered earlier than other vaccines, and recent trials in Indonesia have demonstrated high efficacy. In this study, we use a stochastic individual-based model of rotavirus transmission and disease to estimate the anticipated population-level impact of RV3-BB following delivery according to either an infant (2, 4, 6 months) and neonatal (0, 2, 4 months) schedule. Using our model, which incorporated an age- and household-structured population and estimates of vaccine efficacy derived from trial data, we found both delivery schedules to be effective at reducing infection and disease. We estimated 95-96% reductions in infection and disease in children under 12 months of age when vaccine coverage is 85%. We also estimate high levels of indirect protection from vaccination, including 78% reductions in infection in adults over 17 years of age. Even for lower vaccine coverage of 55%, we estimate reductions of 84% in infection and disease in children under 12 months of age. While open questions remain about the drivers of observed lower efficacy in low-income settings, our model suggests RV3-BB could be effective at reducing infection and preventing disease in young infants at the population level.

Capturing Household Structure and Mobility within and between Remote Aboriginal Communities in Northern Australia Using Longitudinal Data: A Pilot Study.

Cultural practices and development level can influence a population's household structures and mixing patterns. Within some populations, households can be organized across multiple dwellings. This likely affects the spread of infectious disease through these communities; however, current demographic data collection tools do not record these data. METHODS: Between June and October 2018, the Contact And Mobility Patterns in remote Aboriginal Australian communities (CAMP-remote) pilot study recruited Aboriginal mothers with infants in a remote northern Australian community to complete a monthly iPad-based contact survey. RESULTS: Thirteen mother-infant pairs (participants) completed 69 study visits between recruitment and the end of May 2019. Participants reported they and their other children slept in 28 dwellings during the study. The median dwelling occupancy, defined as people sleeping in the same dwelling on the previous night, was ten (range: 3.5-25). Participants who completed at least three responses (n = 8) slept in a median of three dwellings (range: 2-9). Each month, a median of 28% (range: 0-63%) of the participants travelled out of the community. Including these data in disease transmission models amplified estimates of infectious disease spread in the study community, compared to models parameterized using census data. CONCLUSIONS: The lack of data on mixing patterns in populations where households can be organized across dwellings may impact the accuracy of infectious disease models for these communities and the efficacy of public health actions they inform.

Propagation, detection and correction of errors using the sequence database network.

Nucleotide and protein sequences stored in public databases are the cornerstone of many bioinformatics analyses. The records containing these sequences are prone to a wide range of errors, including incorrect functional annotation, sequence contamination and taxonomic misclassification. One source of information that can help to detect errors are the strong interdependency between records. Novel sequences in one database draw their annotations from existing records, may generate new records in multiple other locations and will have varying degrees of similarity with existing records across a range of attributes. A network perspective of these relationships between sequence records, within and across databases, offers new opportunities to detect-or even correct-erroneous entries and more broadly to make inferences about record quality. Here, we describe this novel perspective of sequence database records as a rich network, which we call the sequence database network, and illustrate the opportunities this perspective offers for quantification of database quality and detection of spurious entries. We provide an overview of the relevant databases and describe how the interdependencies between sequence records across these databases can be exploited by network analyses. We review the process of sequence annotation and provide a classification of sources of error, highlighting propagation as a major source. We illustrate the value of a network perspective through three case studies that use network analysis to detect errors, and explore the quality and quantity of critical relationships that would inform such network analyses. This systematic description of a network perspective of sequence database records provides a novel direction to combat the proliferation of errors within these critical bioinformatics resources.

MPVNN: Mutated Pathway Visible Neural Network architecture for interpretable prediction of cancer-specific survival risk.

MOTIVATION: Survival risk prediction using gene expression data is important in making treatment decisions in cancer. Standard neural network (NN) survival analysis models are black boxes with a lack of interpretability. More interpretable visible neural network architectures are designed using biological pathway knowledge. But they do not model how pathway structures can change for particular cancer types. RESULTS: We propose a novel Mutated Pathway Visible Neural Network (MPVNN) architecture, designed using prior signaling pathway knowledge and random replacement of known pathway edges using gene mutation data simulating signal flow disruption. As a case study, we use the PI3K-Akt pathway and demonstrate overall improved cancer-specific survival risk prediction of MPVNN over other similar-sized NN and standard survival analysis methods. We show that trained MPVNN architecture interpretation, which points to smaller sets of genes connected by signal flow within the PI3K-Akt pathway that is important in risk prediction for particular cancer types, is reliable. AVAILABILITY AND IMPLEMENTATION: The data and code are available at https://github.com/gourabghoshroy/MPVNN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Modelling the effect of within-host dynamics on the diversity of a multi-strain pathogen.

Multi-strain pathogens such as Group A Streptococcus, Streptococcus pneumoniae, and Staphylococcus aureus cause millions of infections each year with a substantial health burden. Control of multi-strain pathogens can be complicated by the high strain diversity often observed in endemic settings. It is not well understood how high strain diversity is maintained in populations, given that they compete with each other both directly (within an individual host) and indirectly (via host immunity). Previous modelling studies have investigated how indirect competition affects the prevalence and diversity of strains. However, these studies often make simplifying assumptions about the direct competition that occurs within hosts. Currently, little data is available to validate these assumptions, hence there is a need to clarify how sensitive model outputs are to these assumptions. In this study, we compare the dynamics of multi-strain pathogens under different assumptions about direct competition between strains using an agent-based model. We find that the assumptions made about direct competition can affect the epidemiological dynamics, particularly when there is no long-term immunity following infections and a low rate of importation of non-circulating strains. Our results suggest that while direct and indirect competition can each decrease strain diversity when they act in isolation, they may increase strain diversity when they act together. This finding highlights the importance of examining sensitivity to assumptions about strain competition. In particular, omitting consideration of direct competition can lead to inaccurate estimates of the likely effectiveness of control strategies as changes in strain diversity shift the level of direct strain competition.

The efficacy of sampling strategies for estimating scabies prevalence.

BACKGROUND: Estimating community level scabies prevalence is crucial for targeting interventions to areas of greatest need. The World Health Organisation recommends sampling at the unit of households or schools, but there is presently no standardised approach to scabies prevalence assessment. Consequently, a wide range of sampling sizes and methods have been used. As both prevalence and drivers of transmission vary across populations, there is a need to understand how sampling strategies for estimating scabies prevalence interact with local epidemiology to affect the accuracy of prevalence estimates. METHODS: We used a simulation-based approach to compare the efficacy of different scabies sampling strategies. First, we generated synthetic populations broadly representative of remote Australian Indigenous communities and assigned a scabies status to individuals to achieve a specified prevalence using different assumptions about scabies epidemiology. Second, we calculated an observed prevalence for different sampling methods and sizes. RESULTS: The distribution of prevalence in subpopulation groups can vary substantially when the underlying scabies assignment method changes. Across all of the scabies assignment methods combined, the simple random sampling method produces the narrowest 95% confidence interval for all sample sizes. The household sampling method introduces higher variance compared to simple random sampling when the assignment of scabies includes a household-specific component. The school sampling method overestimates community prevalence when the assignment of scabies includes an age-specific component. DISCUSSION: Our results indicate that there are interactions between transmission assumptions and surveillance strategies, emphasizing the need for understanding scabies transmission dynamics. We suggest using the simple random sampling method for estimating scabies prevalence. Our approach can be adapted to various populations and diseases.

Exploring automatic inconsistency detection for literature-based gene ontology annotation.

MOTIVATION: Literature-based gene ontology annotations (GOA) are biological database records that use controlled vocabulary to uniformly represent gene function information that is described in the primary literature. Assurance of the quality of GOA is crucial for supporting biological research. However, a range of different kinds of inconsistencies in between literature as evidence and annotated GO terms can be identified; these have not been systematically studied at record level. The existing manual-curation approach to GOA consistency assurance is inefficient and is unable to keep pace with the rate of updates to gene function knowledge. Automatic tools are therefore needed to assist with GOA consistency assurance. This article presents an exploration of different GOA inconsistencies and an early feasibility study of automatic inconsistency detection. RESULTS: We have created a reliable synthetic dataset to simulate four realistic types of GOA inconsistency in biological databases. Three automatic approaches are proposed. They provide reasonable performance on the task of distinguishing the four types of inconsistency and are directly applicable to detect inconsistencies in real-world GOA database records. Major challenges resulting from such inconsistencies in the context of several specific application settings are reported. This is the first study to introduce automatic approaches that are designed to address the challenges in current GOA quality assurance workflows. The data underlying this article are available in Github at https://github.com/jiyuc/AutoGOAConsistency.

COVID-19 in low-tolerance border quarantine systems: Impact of the Delta variant of SARS-CoV-2.

In controlling transmission of coronavirus disease 2019 (COVID-19), the effectiveness of border quarantine strategies is a key concern for jurisdictions in which the local prevalence of disease and immunity is low. In settings like this such as China, Australia, and New Zealand, rare outbreak events can lead to escalating epidemics and trigger the imposition of large-scale lockdown policies. Here, we develop and apply an individual-based model of COVID-19 to simulate case importation from managed quarantine under various vaccination scenarios. We then use the output of the individual-based model as input to a branching process model to assess community transmission risk. For parameters corresponding to the Delta variant, our results demonstrate that vaccination effectively counteracts the pathogen's increased infectiousness. To prevent outbreaks, heightened vaccination in border quarantine systems must be combined with mass vaccination. The ultimate success of these programs will depend sensitively on the efficacy of vaccines against viral transmission.

Rapid assessment of the risk of SARS-CoV-2 importation: case study and lessons learned.

During the early stages of an emerging disease outbreak, governments are required to make critical decisions on how to respond, despite limited data being available to inform these decisions. Analytical risk assessment is a valuable approach to guide decision-making on travel restrictions and border measures during the early phase of an outbreak. Here we describe a rapid risk assessment framework that was developed in February 2020 to support time-critical decisions on the risk of SARS-CoV-2 importation into Australia. We briefly describe the context in which our framework was developed, the framework itself, and provide an example of the type of decision support provided to the Australian government. We then report a critical evaluation of the modelling choices made in February 2020, assessing the impact of our assumptions on estimated rates of importation, and provide a summary of "lessons learned". The framework presented and evaluated here provides a flexible approach to rapid assessment of importation risk, of relevance to current and future pandemic scenarios.

Automatic consistency assurance for literature-based gene ontology annotation.

BACKGROUND: Literature-based gene ontology (GO) annotation is a process where expert curators use uniform expressions to describe gene functions reported in research papers, creating computable representations of information about biological systems. Manual assurance of consistency between GO annotations and the associated evidence texts identified by expert curators is reliable but time-consuming, and is infeasible in the context of rapidly growing biological literature. A key challenge is maintaining consistency of existing GO annotations as new studies are published and the GO vocabulary is updated. RESULTS: In this work, we introduce a formalisation of biological database annotation inconsistencies, identifying four distinct types of inconsistency. We propose a novel and efficient method using state-of-the-art text mining models to automatically distinguish between consistent GO annotation and the different types of inconsistent GO annotation. We evaluate this method using a synthetic dataset generated by directed manipulation of instances in an existing corpus, BC4GO. We provide detailed error analysis for demonstrating that the method achieves high precision on more confident predictions. CONCLUSIONS: Two models built using our method for distinct annotation consistency identification tasks achieved high precision and were robust to updates in the GO vocabulary. Our approach demonstrates clear value for human-in-the-loop curation scenarios.

Mapping home internet activity during COVID-19 lockdown to identify occupation related inequalities.

During the COVID-19 pandemic, evidence has accumulated that movement restrictions enacted to combat virus spread produce disparate consequences along socioeconomic lines. We investigate the hypothesis that people engaged in financially secure employment are better able to adhere to mobility restrictions, due to occupational factors that link the capacity for flexible work arrangements to income security. We use high-resolution spatial data on household internet traffic as a surrogate for adaptation to home-based work, together with the geographical clustering of occupation types, to investigate the relationship between occupational factors and increased internet traffic during work hours under lockdown in two Australian cities. By testing our hypothesis based on the observed trends, and exploring demographic factors associated with divergences from our hypothesis, we are left with a picture of unequal impact dominated by two major influences: the types of occupations in which people are engaged, and the composition of households and families. During lockdown, increased internet traffic was correlated with income security and, when school activity was conducted remotely, to the proportion of families with children. Our findings suggest that response planning and provision of social and economic support for residents within lockdown areas should explicitly account for income security and household structure. Overall, the results we present contribute to the emerging picture of the impacts of COVID-19 on human behaviour, and will help policy makers to understand the balance between public health and social impact in making decisions about mitigation policies.