Text Mining Drug/Chemical-Protein Interactions using an Ensemble of BERT and T5 Based Models

TL;DR

This paper compares BERT-based and T5-based models for extracting drug/chemical-protein interactions, finding BERT models perform better but T5 shows promising potential for future research.

Contribution

Introduces a novel T5 text-to-text approach for relation extraction and compares it with traditional BERT-based models in drug-chemical protein interaction prediction.

Findings

BERT-based BioMegatron model achieved 0.74 F1 score.

T5 model achieved 0.65 F1 score, outperforming similar data-trained models.

Larger BERT models generally perform better in this task.

Abstract

In Track-1 of the BioCreative VII Challenge participants are asked to identify interactions between drugs/chemicals and proteins. In-context named entity annotations for each drug/chemical and protein are provided and one of fourteen different interactions must be automatically predicted. For this relation extraction task, we attempt both a BERT-based sentence classification approach, and a more novel text-to-text approach using a T5 model. We find that larger BERT-based models perform better in general, with our BioMegatron-based model achieving the highest scores across all metrics, achieving 0.74 F1 score. Though our novel T5 text-to-text method did not perform as well as most of our BERT-based models, it outperformed those trained on similar data, showing promising results, achieving 0.65 F1 score. We believe a text-to-text approach to relation extraction has some competitive advantages and there is a lot of room for research advancement.