BioBLP: A Modular Framework for Learning on Multimodal Biomedical Knowledge Graphs

TL;DR

BioBLP introduces a modular framework for learning embeddings on heterogeneous biomedical knowledge graphs, effectively incorporating diverse attribute data and enabling efficient pretraining for improved link and interaction predictions.

Contribution

It presents a novel modular approach that handles multiple attribute modalities and missing data, along with an efficient pretraining strategy for biomedical KGs.

Findings

Competitive link prediction performance, slightly lower than baselines without attributes.

Outperforms baselines in drug-protein interaction prediction.

Pretraining strategy significantly improves performance and reduces training time.

Abstract

Knowledge graphs (KGs) are an important tool for representing complex relationships between entities in the biomedical domain. Several methods have been proposed for learning embeddings that can be used to predict new links in such graphs. Some methods ignore valuable attribute data associated with entities in biomedical KGs, such as protein sequences, or molecular graphs. Other works incorporate such data, but assume that entities can be represented with the same data modality. This is not always the case for biomedical KGs, where entities exhibit heterogeneous modalities that are central to their representation in the subject domain. We propose a modular framework for learning embeddings in KGs with entity attributes, that allows encoding attribute data of different modalities while also supporting entities with missing attributes. We additionally propose an efficient pretraining strategy for reducing the required training runtime. We train models using a biomedical KG containing approximately 2 million triples, and evaluate the performance of the resulting entity embeddings on the tasks of link prediction, and drug-protein interaction prediction, comparing against methods that do not take attribute data into account. In the standard link prediction evaluation, the proposed method results in competitive, yet lower performance than baselines that do not use attribute data. When evaluated in the task of drug-protein interaction prediction, the method compares favorably with the baselines. We find settings involving low degree entities, which make up for a substantial amount of the set of entities in the KG, where our method outperforms the baselines. Our proposed pretraining strategy yields significantly higher performance while reducing the required training runtime. Our implementation is available at https://github.com/elsevier-AI-Lab/BioBLP .