Artificial Neural Network Prediction of Viruses in Shellfish

Abstract

A database was probed with artificial neural network (ANN) and multivariate logistic regression (MLR) models to investigate the efficacy of predicting PCR-identified human adenovirus (ADV), Norwalk-like virus (NLV), and enterovirus (EV) presence or absence in shellfish harvested from diverse countries in Europe (Spain, Sweden, Greece, and the United Kingdom). The relative importance of numerical and heuristic input variables to the ANN model for each country and for the combined data was analyzed with a newly defined relative strength effect, which illuminated the importance of bacteriophages as potential viral indicators. The results of this analysis showed that ANN models predicted all types of viral presence and absence in shellfish with better precision than MLR models for a multicountry database. For overall presence/absence classification accuracy, ANN modeling had a performance rate of 95.9%, 98.9%, and 95.7% versus 60.5%, 75.0%, and 64.6% for the MLR for ADV, NLV, and EV, respectively. The selectivity (prediction of viral negatives) was greater than the sensitivity (prediction of viral positives) for both models and with all virus types, with the ANN model performing with greater sensitivity than the MLR. ANN models were able to illuminate site-specific relationships between microbial indicators chosen as model inputs and human virus presence. A validation study on ADV demonstrated that the MLR and ANN models differed in sensitivity and selectivity, with the ANN model correctly identifying ADV presence with greater precision.Health risks associated with the consumption of virally contaminated shellfish are well documented, as is the need for a more reliable viral indicator system by the industry (14, 17). Interdisciplinary studies are needed to define the underlying relationships between harvest area water quality, shellfish type, treatment processes, and viral presence, particularly with the advent of advanced detection and modeling methods. New understandings can be obtained with the application of new data-driven, fuzzy-logic-based models that can handle multiple, interrelated inputs and learn complex relationships. However, for the application of these new models, large, robust, multivariable, complete, and well-controlled datasets need to be created. A multicountry study in Europe has collected vital data in an effort to relate the viral contamination of shellfish with potential indicators. Analysis of these results has been reported earlier by 7), and the database consisted of 468 individual observations from geographically diverse areas collected over 18 months. The resultant database was provided to a team of engineers and modeling experts for further probing with new artificial neural network (ANN) modeling tools under the hypothesis that these new modeling tools would be able to better define the relationships between viral presence/absence and potential water quality indicators than multivariate logistic regression (MLR) and provide more precise predictions with ANN models.Neural network model...

Neural Network Analysis to Evaluate Ozone Damage to Vegetation Under Different Climatic Conditions

“…Our results are in line with other ecological studies in literature which compared the two methods and found ANN models more accurate than MLR (e.g. Lek et al, 1996;Paruelo and Tomasel, 1997;Brion et al, 2005). The strength of ANN lies in its fully inductive approach, which allows multidimensional relationships to be investigated without a priori knowledge of the shape of these relations.…”

Risk management of shellfisheries

“…Work is currently ongoing within CEN to produce a standard method based on this technology. Brion et al (2005) showed that, for three European countries, the occurrence of norovirus varied significantly between the countries and that additionally, presence was associated with area within the country, month, bacteriophage (somatic coliphage, F+-bacteriophage and B. fragilis phage) concentrations and species of mollusc being important. The relative importance of factors varied among the countries ± it may therefore be necessary to undertake regional, national or local risk profiles when determining the factors to be included in risk management strategies.…”

Programmable DNA-Based Boolean Logic Microfluidic Processing Unit

“…Deoxyribonucleic acid (DNA) is considered as a promising material for molecular computing, which involves logical calculations , and solving complex mathematical problems − based on the chemical reactions between molecules. Owing to Watson–Crick pairing, DNA information binding with paired sequences is widely used in traditional biotechnologies such as polymer chain reaction (PCR), gene editing and synthesis, , and molecule sensing, − as well as for bionanotechnologies such as superlattices, DNA origami, and DNA computing. , The chemical reaction network (CRN), which is conducted through parallel and cascading reactions in the DNA solution enables solving problems via DNA biocomputing, which includes logic gate calculations, neural networks, , cell computing, and virus detection using neural networks. , However, several sequential CRN steps are required for implementing simple DNA-based logic gates; these steps are typically demonstrated in a single, small PCR tube or a plate reader by adding molecules using a pipet. , For example, Qian et al . designed CRNs for basic logic gates such as the AND, OR, and NOT gates by using four or five sequential steps .…”

“…14,15 The chemical reaction network (CRN), which is conducted through parallel and cascading reactions in the DNA solution enables solving problems via DNA biocomputing, which includes logic gate calculations, 16 neural networks, 17,18 cell computing, 19 and virus detection using neural networks. 20,21 these steps are typically demonstrated in a single, small PCR tube or a plate reader by adding molecules using a pipet. 22,23 For example, Qian et al designed CRNs for basic logic gates such as the AND, OR, and NOT gates by using four or five sequential steps.…”