Gradient boosting machine learning model to predict aflatoxins in Iowa corn

Authors: BRANSTAD-SPATES, EMILY - Iowa State University; Castano-Duque, Lina; MOSHER, GRETCHEN - Iowa State University; HURBURGH, JR., CHARLES - Iowa State University; Owens, Phillip; Winzeler, Hans - Edwin; Rajasekaran, Kanniah - Rajah; BOWERS, ERIN - Iowa State University

Submitted to: Frontiers in Microbiology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/14/2023
Publication Date: 9/1/2023
Citation: Branstad-Spates, E.H., Castano-Duque, L., Mosher, G.A., Hurburgh, Jr., C.R., Owens, P., Winzeler, E., Rajasekaran, K., Bowers, E.L. 2023. Gradient boosting machine learning model to predict aflatoxins in Iowa corn. Frontiers in Microbiology. 14:1248772. https://doi.org/10.3389/fmicb.2023.1248772.
DOI: https://doi.org/10.3389/fmicb.2023.1248772

Interpretive Summary: Aflatoxin (AFL), a secondary metabolite produced from filamentous fungi, contaminates corn, posing significant health and safety hazards for humans and livestock due to their toxigenic and carcinogenic effects. In this case study, an Iowa-centric Machine Learning model was developed to predict AFL contamination using historical corn contamination, meteorological data, satellite data, and soil property data in the largest corn-producing state in the US. Analyses for AFL showed that satellite-acquired vegetative index data during August significantly influenced corn contamination at the end of growing season. Prediction of high AFL contamination levels was also linked to aflatoxin risk indices (ARI, which includes all the influential parameters) in May. Developing these AFL prediction models is practical and implementable in commodity grain handling environments to prevent contamination. Finding factors that influence AFL risk thresholds each year will serve as an effective tool to increase food and feed safety. These findings will be useful to all growers and stakeholders in the corn industry.

Technical Abstract: Aflatoxin (AFL), a secondary metabolite produced from filamentous fungi, contaminates corn, posing significant health and safety hazards for humans and livestock due to their toxigenic and carcinogenic effects. Corn is widely used as an essential commodity for food, feed, fuel, and export markets; therefore, AFL mitigation is necessary to ensure food and feed safety within the United States (US) and elsewhere in the world. In this case study, an Iowa-centric model was developed to predict AFL contamination using historical corn contamination, meteorological data, satellite data, and soil property data in the largest corn-producing state in the US. We evaluated the performance of AFL prediction with gradient boosting machine (GBM) learning and feature engineering in Iowa corn for two AFL risk thresholds for high contamination events: 20-ppb and 5-ppb. The GBM model had an overall accuracy of 96.77% for AFL with a balanced accuracy of 50.00% for a 20-ppb risk threshold, whereas GBM had an overall accuracy of 90.32% with a balanced accuracy of 64.88% for a 5-ppb threshold. Analyses for AFL showed that satellite-acquired vegetative index data during August significantly influenced corn contamination at the end of the growing season for both risk thresholds. Prediction of high AFL contamination levels was linked to aflatoxin risk indices (ARI) in May. However, ARI in July was an influential factor for the 5-ppb threshold but not for the 20-ppb threshold. Similarly, latitude was an influential factor for the 20-ppb threshold but not the 5-ppb threshold. Furthermore, soil-saturated hydraulic conductivity (Ksat) influenced both risk thresholds. Developing these AFL prediction models is practical and implementable in commodity grain handling environments to achieve the goal of preventative rather than reactive mitigations. Finding factors that influence AFL risk thresholds each year is an important cost-effective risk tool and, therefore, is a high priority to ensure hazard management and optimal grain utilization to maximize the utility and value of the nation’s corn crop.