Machine Learning FLIT-SHAP Reveal key contaminant interactions to enhance toxicity prediction and environmental health decision-making
Busan, South Korea, July 23, 2024 /PRNewswire/ — Traditional environmental health studies often focus on the toxicity of exposure to a single chemical. However, in the real world, people are exposed to multiple contaminants simultaneously, which may interact in complex ways to amplify or diminish toxic effects. Traditional models that assume additive effects, such as concentration addition or independent action, can be misleading in such scenarios. Advanced statistical and machine learning methods have been employed to address this issue, but they are often inadequate due to the complexity, the large number of interacting contaminants, and the inability to extract the absolute effect of each contaminant.
The new FLIT-SHAP approach helps to reveal hidden interactions between contaminants, showing both synergistic and antagonistic effects on toxicity with an R² of 0.99.
To address this issue, a group of researchers led by Professor Kuk Cho From Busan National university Feature Local Intercept Transform Shapley Additive Explanation (FLIT-SHAP) was introduced as a solution to these challenges. This tool is unique in that it analyzes the impact of specific pollutants within a mixture, unlike traditional methods that use a broader approach. This detail-oriented technique is particularly useful for understanding the OPs of airborne particulate matter (PM), which is known to cause a variety of health problems. The study Hazardous Materials Journalupon August 14, 2024.
“The OPs in PM refer to the ability of these particles to generate reactive oxygen species, which can cause oxidative stress and inflammation in the body, leading to a variety of health problems, including respiratory and cardiovascular diseases.“The scientists used FLIT-SHAP to study the OPs of these particles in contaminant mixtures. The lab-curated data was used to train robust machine learning models such as eXtreme Gradient Boosting (XGBoost), and the FLIT-SHAP explanatory tool was used to reveal key insights.
“In controlled laboratory environments, we found that 55-63% of interactions were synergistic (working together to increase toxicity) and 25-42% were antagonistic (counteracting each other). However, in real-world scenarios, antagonistic effects are more common, resulting in lower overall toxicity than predicted by previous models.Professor Cho added.
One surprising finding was the significant impact of a group of chemicals called quinones, which contributes more to toxicity than previously thought. This finding calls into question current approaches to regulating pollutants and suggests that more focus should be placed on these chemicals. FLIT-SHAP not only predicted toxicity with excellent accuracy (R² = 0.99), but also provided detailed insight into how different contaminants interact with each other. This makes it a powerful tool for conducting realistic risk assessments and improving our understanding of environmental health risks.
The researchers' rigorous testing revealed how contaminants interact at different concentrations and revealed that traditional additive models often over- or underestimate the toxicity of mixtures. Through FLIT-SHAP, the researchers identified important nuances in the contribution of contaminants in different environments, highlighting the need for more refined risk assessments.
“While this study focuses on specific contaminants and toxicity endpoints, further research on broader scenarios is essential. FLIT-SHAP represents a major leap forward, providing an accurate tool for assessing the toxicity of chemical mixtures. This advance promises better regulatory decisions to effectively protect human health.”
reference
Original paper title: Machine learning-based dose-response relationships accounting for interactions in mixtures: Application to the oxidation potential of particulate matter
journal: Hazardous Materials Journal
Source: https://doi.org/10.1016/j.jhazmat.2024.134864
About the Institute
Website: https://www.pusan.ac.kr/eng/Main.do
contact:
Lee Jae Eun
82 51 510 7928
[email protected]
Source: Busan National university
