Prediction of adverse biological effects of chemicals using knowledge graph embeddings

Abstract

We have created a knowledge graph based on major data sources used in ecotoxicological risk assessment. We have applied this knowledge graph to an important task in risk assessment, namely chemical effect prediction. We have evaluated nine knowledge graph embedding models from a selection of geometric, decomposition, and convolutional models on this prediction task. We show that using knowledge graph embeddings can increase the accuracy of effect prediction with neural networks. Furthermore, we have implemented a fine-tuning architecture which adapts the knowledge graph embeddings to the effect prediction task and leads to a better performance. Finally, we evaluate certain characteristics of the knowledge graph embedding models to shed light on the individual model performance.

Editorial of the Special Issue on Deep Learning and Knowledge Graphs

“…The eleventh paper, "Prediction of Adverse Biological Effects of Chemicals Using Knowledge Graph Embeddings" [13] by Erik Bryhn Myklebust, Ernesto Jimenez-Ruiz, Jiaoyan Chen, Raoul Wolf, and Knut Erik Tollefsen. The paper proposes a KG based on major data sources used in ecotoxicological risk assessment.…”

A Few-Shot Inductive Link Prediction Model in Knowledge Graphs

“…Representing facts form reality in from of the triplet (head entity, relationship, tail entity) establishing the relationship between entities is the most typical representation method. In recent years, knowledge graphs already played an important role in supporting for information retrieval [1], intelligent question answering [2], [3] recommendation system [4], and other fields related to artificial intelligence [5], [6]. However, most constructed knowledge graphs are incomplete due to the limitations of existing knowledge and extraction algorithms.…”