Understanding Adverse Biological Effect Predictions Using Knowledge Graphs
This work addresses the need for more efficient hazard and risk assessment in toxicology by reducing reliance on animal experiments, though it appears incremental as it builds on existing prediction models with added knowledge graphs.
The paper tackled the problem of predicting adverse biological effects of chemicals without animal testing by using a knowledge graph as background knowledge, resulting in a 40% improvement in prediction performance measured by R^2.
Extrapolation of adverse biological (toxic) effects of chemicals is an important contribution to expand available hazard data in (eco)toxicology without the use of animals in laboratory experiments. In this work, we extrapolate effects based on a knowledge graph (KG) consisting of the most relevant effect data as domain-specific background knowledge. An effect prediction model, with and without background knowledge, was used to predict mean adverse biological effect concentration of chemicals as a prototypical type of stressors. The background knowledge improves the model prediction performance by up to 40\% in terms of $R^2$ (\ie coefficient of determination). We use the KG and KG embeddings to provide quantitative and qualitative insights into the predictions. These insights are expected to improve the confidence in effect prediction. Larger scale implementation of such extrapolation models should be expected to support hazard and risk assessment, by simplifying and reducing testing needs.