ATQ: alert time quality, an evaluation metric for assessing timely epidemic detection models within a school absenteeism-based surveillance system.

BACKGROUND: Wellington-Dufferin-Guelph Public Health (WDGPH) has conducted an absenteeism-based influenza surveillance program in the WDG region of Ontario, Canada since 2008, using a 10% absenteeism threshold to raise an alert for the implementation of mitigating measures. A recent study indicated that model-based alternatives, such as distributed lag seasonal logistic regression models, provided improved alerts for detecting an upcoming epidemic. However model evaluation and selection was primarily based on alert accuracy, measured by the false alert rate (FAR), and failed to optimize timeliness. Here, a new metric that simultaneously evaluates epidemic alert accuracy and timeliness is proposed. The alert time quality (ATQ) metric is investigated as a model selection criterion on both a simulated and real data set. METHODS: The ATQ assessed alerts on a gradient, where alerts raised incrementally before or after an optimal day were considered informative, but were penalized for lack of timeliness. Summary statistics of ATQ, average alert time quality (AATQ) and first alert time quality (FATQ), were used for model evaluation and selection. Alerts raised by ATQ and FAR selected models were compared. Daily elementary school absenteeism and laboratory-confirmed influenza case data collected by WDGPH were used for demonstration and evaluation of the proposed metric. A simulation study that mimicked the WDG population and influenza demographics was conducted for further evaluation of the proposed metric. RESULTS: The FATQ-selected model raised acceptable first alerts most frequently, while the AATQ-selected model raised first alerts within the ideal range most frequently. CONCLUSIONS: Models selected by either FATQ or AATQ would more effectively predict community influenza activity with the local community than those selected by FAR.

The acute toxicity of bitumen-influenced groundwaters from the oil sands region to aquatic organisms.

The extraction of surface mined bitumen from oil sands deposits in northern Alberta, Canada produces large quantities of liquid tailings waste, termed oil sands process-affected water (OSPW), which are stored in large tailings ponds. OSPW-derived chemicals from several tailings ponds migrating past containment structures and through groundwater systems pose a concern for surface water contamination. The present study investigated the toxicity of groundwater from near-field sites adjacent to a tailings pond with OPSW influence and far-field sites with only natural oil sands bitumen influence. The acute toxicity of unfractionated groundwater and isolated organic fractions was assessed using a suite of aquatic organisms (Pimephales promelas, Oryzias latipes, Daphnia magna, Hyalella azteca, Lampsilis spp., Ceriodaphnia dubia, Hexagenia spp., and Vibrio fischeri). Assessment of unfractionated groundwater demonstrated toxicity towards all invertebrates in at least one far-field sample, with both near-field and far-field samples with bitumen influence toxic towards P. promelas, while no toxicity was observed for O. latipes. When assessing the unfractionated groundwater and isolated organic fractions from near-field and far-field groundwater sites, P. promelas and H. azteca were the most sensitive to organic components, while D. magna and L. cardium were most sensitive to the inorganic components. Groundwater containing appreciable amounts of dissolved organics exhibited similar toxicities to sensitive species regardless of an OSPW or natural bitumen source. The lack of a clear distinction in relative acute toxicities between near-field and far-field samples indicates that the water-soluble chemicals associated with bitumen are acutely toxic to several aquatic organisms. This result, combined with the similarities in chemical profiles between bitumen-influenced groundwater originating from OSPW and/or natural sources, suggests that the industrial bitumen extraction processes corresponding to the tailings pond in this study are not contributing unique toxic substances to groundwater, relative to natural bitumen compounds present in groundwater flow systems.

Computationally efficient parameter estimation for spatial individual-level models of infectious disease transmission.

Individual-level models incorporate individual-specific covariate information, such as spatial location, to model infectious disease transmission. However, fitting these models with traditional Bayesian methods becomes cumbersome as model complexity or population size increases. We consider a spatial individual-level model with a binary susceptibility covariate. A method for fitting this model to aggregate-level data using traditional Metropolis-Hastings MCMC and then disaggregating the results to obtain individual-level estimates for epidemic metrics is proposed. This so-called "Cluster-Aggregate-Disaggregate" (CAD) method is compared to two approximate Bayesian computation (ABC) algorithms in a simulation study. The methods are also applied to a data set from the 2001 U.K. foot and mouth disease epidemic. While the CAD and ABC methods both performed reasonably well at capturing epidemic metrics, the CAD method was found to be much easier to implement and reduced computation time (relative to the traditional model-fitting method) more consistently than the ABC methods.

Improving monitoring of fish health in the oil sands region using regularization techniques and water quality variables.

Trout-perch are sampled from the Athabasca River in Alberta, Canada, as a sentinel species for environmental health. The performance of trout-perch populations is known to be influenced by the quality of the water in which they reside. Using climate, environmental, and water quality variables measured in the Athabasca River near trout-perch sampling locations is found to improve model fitting and the predictability of models for the adjusted body weight, adjusted gonad weight, and adjusted liver weight of trout-perch. Given a large number of covariables, three variable selection techniques: stepwise regression, the lasso, and the elastic net (EN) are considered for selecting a subset of relevant variables. The models selected by the lasso and EN are found to outperform the models selected by stepwise regression in general, and little difference is observed between the models selected by the lasso and EN. Uranium, tungsten, tellurium, pH, molybdenum, and antimony are selected for at least one fish response.

Correlation Between KIT Expression and c-Kit Mutations in 2 Subtypes of Canine Oral Melanocytic Neoplasms.

c-Kit mutations have been reported in 15% to 40% of certain human melanoma subtypes, including those histologically similar to canine oral malignant melanomas. Therapeutic response to tyrosine kinase inhibitors has been demonstrated in those human patients. As canine oral malignant melanomas tend to have a poor prognosis despite aggressive surgical removal, evaluation of KIT expression and identification of c-Kit mutations in canine oral melanocytic neoplasms was performed to determine if there is any indication that tyrosine kinase inhibitor drugs might effectively treat any of these cases. This study evaluated 27 canine oral malignant melanomas and 12 canine histologically well-differentiated oral melanocytic neoplasms for activating c-Kit mutations, determined differences in immunohistochemical expression of KIT and c-Kit mutation status, and determined if KIT expression could predict c-Kit mutation status. Among samples that contained intraepithelial nests of neoplastic melanocytes in the KIT-labeled sections, KIT was expressed within cells in these nests in 22/23 (96%) malignant melanomas and 5/7 histologically well-differentiated neoplasms. KIT was expressed in 10% to 30% of neoplastic melanocytes in the lamina propria in 3/24 (13%) malignant melanomas, but 0/9 (0%) histologically well-differentiated neoplasms. Next-generation sequencing identified 85 variants in c-Kit, including 9 nonsynonymous mutations that resulted in amino acid changes predicted to affect protein function. c-Kit mutations with predicted deleterious protein effects were more common in malignant melanomas (8/27 [30%] vs 1/12 [8%]). There was no apparent relationship between detected c-Kit mutations and KIT expression. These results do not support the use of therapies that target c-Kit.

Molecular Detection Mapping and Analysis Platform for R (MDMAPR) facilitating the standardization, analysis, visualization, and sharing of qPCR data and metadata.

Quantitative polymerase chain reaction (qPCR) has been used as a standard molecular detection tool in many scientific fields. Unfortunately, there is no standard method for managing published qPCR data, and those currently used generally focus on only managing raw fluorescence data. However, associated with qPCR experiments are extensive sample and assay metadata, often under-examined and under-reported. Here, we present the Molecular Detection Mapping and Analysis Platform for R (MDMAPR), an open-source and fully scalable informatics tool for researchers to merge raw qPCR fluorescence data with associated metadata into a standard format, while geospatially visualizing the distribution of the data and relative intensity of the qPCR results. The advance of this approach is in the ability to use MDMAPR to store varied qPCR data. This includes pathogen and environmental qPCR species detection studies ideally suited to geographical visualization. However, it also goes beyond these and can be utilized with other qPCR data including gene expression studies, quantification studies used in identifying health dangers associated with food and water bacteria, and the identification of unknown samples. In addition, MDMAPR's novel centralized management and geospatial visualization of qPCR data can further enable cross-discipline large-scale qPCR data standardization and accessibility to support research spanning multiple fields of science and qPCR applications.

Development of a Reduced-Volume Acute Lethality Toxicity Test for Hyalella azteca.

Effects-directed analysis (EDA) is used to identify the principal toxic components within a complex mixture using iterative steps of chemical fractionation guided by bioassay results. Bioassay selection can be limited in EDA because of the volume requirements for many standardized test methods, and therefore, a reduced-volume acute toxicity test that also provides whole-organism responses is beneficial. To address this need, a static, 7-d, water-only, reduced-volume method (50 mL, 10 organisms) was developed for Hyalella azteca that substantially decreases the volume requirements of standard-volume acute test exposures (200-500 mL of test solution, 15-20 organisms) while maintaining water quality and meeting control survival criteria. Standard- and reduced-volume methods were compared by conducting concurrent toxicity tests with 2 inorganic toxicants (KCl and CdCl2 ) and 2 organic mixtures of naphthenic acid fraction components (NAFCs) to evaluate test performance. There was no difference between methods when comparing the median lethal concentrations (LC50s) for KCl and both NAFC mixtures (p > 0.05). The LC50s for CdCl2 were statistically different (p = 0.0002); however, this was not considered biologically meaningful because the difference between LC50s was <2-fold. In conclusion, the reduced-volume H. azteca test method generated results comparable to standard-volume test methods and is suitable for use in situations where limited testing material is available, such as when conducting EDA. Environ Toxicol Chem 2020;39:2221-2227. © Her Majesty the Queen in Right of Canada 2020. Reproduced with the permission of the Minister of Environment and Climate Change Canada.

Methods for detecting seasonal influenza epidemics using a school absenteeism surveillance system.

BACKGROUND: School absenteeism data have been collected daily by the public health unit in Wellington-Dufferin-Guelph, Ontario since 2008. To date, a threshold-based approach has been implemented to raise alerts for community-wide and within-school illness outbreaks. We investigate several statistical modelling approaches to using school absenteeism for influenza surveillance at the regional level, and compare their performances using two metrics. METHODS: Daily absenteeism percentages from elementary and secondary schools, and report dates for influenza cases, were obtained from Wellington-Dufferin-Guelph Public Health. Several absenteeism data aggregations were explored, including using the average across all schools or only using schools of one type. A 10% absence threshold, exponentially weighted moving average model, logistic regression with and without seasonality terms, day of week indicators, and random intercepts for school year, and generalized estimating equations were used as epidemic detection methods for seasonal influenza. In the regression models, absenteeism data with various lags were used as predictor variables, and missing values in the datasets used for parameter estimation were handled either by deletion or linear interpolation. The epidemic detection methods were compared using a false alarm rate (FAR) as well as a metric for alarm timeliness. RESULTS: All model-based epidemic detection methods were found to decrease the FAR when compared to the 10% absence threshold. Regression models outperformed the exponentially weighted moving average model and including seasonality terms and a random intercept for school year generally resulted in fewer false alarms. The best-performing model, a seasonal logistic regression model with random intercept for school year and a day of week indicator where parameters were estimated using absenteeism data that had missing values linearly interpolated, produced a FAR of 0.299, compared to the pre-existing threshold method which at best gave a FAR of 0.827. CONCLUSIONS: School absenteeism can be a useful tool for alerting public health to upcoming influenza epidemics in Wellington-Dufferin-Guelph. Logistic regression with seasonality terms and a random intercept for school year was effective at maximizing true alarms while minimizing false alarms on historical data from this region.

Assessing the bioremediation potential of algal species indigenous to oil sands process-affected waters on mixtures of oil sands acid extractable organics.

Surface mining extraction of bitumen from oil sand in Alberta, Canada results in the accumulation of oil sands process-affected water (OSPW). In attempts to maximize water recycling, and because its constituents are recognized as being toxic, OSPW is retained in settling basins. Consequently, research efforts are currently focused on developing remediation strategies capable of detoxifying OSPW to allow for eventual release. One potential bioremediation strategy proposes to utilize phytoplankton native to the Alberta oil sand region to sequester, break down, or modify the complex oil sands acid extractable organic (AEO) mixtures in OSPW. Preliminary attempts to quantify changes in total oil sands AEO concentration in test solutions by ESI-MS following a 14-day algal remediation period revealed the presence of unknown organic acids in control samples, likely released by the phytoplankton strains and often of the same atomic mass range as the oil sands AEO under investigation. To address the presence of these "biogenic" organic acids in test samples, ESI-MS in MRM mode was utilized to identify oil sands AEO "marker ions" that were a) present within the tested oil sands AEO extract and b) unique to the oil sands AEO extract only (e.g. atomic masses different from biogenic organic acids). Using this approach, one of the 21 tested algal strains, Stichococcus sp. 1, proved capable of significantly reducing the AEO marker ion concentration at test concentrations of 10, 30, and 100mgL(-1). This result, along with the accelerated growth rate and recalcitrance of this algal strain with exposure to oil sands AEO, suggests the strong potential for the use of the isolated Stichococcus sp. 1 as a candidate for bioremediation strategies.

Spatial data aggregation for spatio-temporal individual-level models of infectious disease transmission.

A class of complex statistical models, known as individual-level models, have been effectively used to model the spread of infectious diseases. These models are often fitted within a Bayesian Markov chain Monte Carlo framework, which can have a sig nificant computational expense due to the complex nature of the likelihood function associated with this class of models. Increases in population size or duration of the modeled epidemic can contribute to this computational burden. Here, we explore the effect of reducing this computational expense by aggregating the data into spatial clusters, and therefore reducing the overall population size. Individual-level models, reparameterized to account for this aggregation effect, may then be fitted to the spatially aggregated data. The ability of two reparameterized individual-level models, when fitted to this reduced data set, to identify a covariate effect is investigated through a simulation study.

Latent conditional individual-level models for infectious disease modeling.

Individual-level models (ILMs) have previously been used to model the spatiotemporal spread of infectious diseases. These models can incorporate individual-level covariate information, to account for population heterogeneity. However, incomplete or unreliable data are a common problem in infectious disease modeling, and models that are explicitly dependent on such information may not be robust to these inherent uncertainties. In this investigation, we assess an adaptation to a spatial ILM that incorporates a latent grouping structure based on some trait heterogeneous in the population. The resulting latent conditional ILM is then only dependent upon a discrete latent grouping variable, rather than precise covariate information. The posterior predictive ability of this proposed model is tested through a simulation study, in which the model is fitted to epidemic data simulated from a true model that utilizes explicit covariate information. In addition, the posterior predictive ability of the proposed ILM is also compared to that of an ILM that assumes population homogeneity. The application of these models to data from the 2001 UK foot-and-mouth disease epidemic is also explored. This study demonstrates that the use of a discrete latent grouping variable can be an effective alternative to utilizing covariate information, particularly when such information may be unreliable.