A bioactivity foundation model using pairwise meta-learning

Abstract

The bioactivity of compounds plays an important role in drug development and discovery. Existing machine learning approaches have poor generalizability in bioactivity prediction due to the small number of compounds in each assay and incompatible measurements among assays. In this paper, we propose ActFound, a bioactivity foundation model trained on 1.6 million experimentally measured bioactivities and 35,644 assays from ChEMBL. The key idea of ActFound is to use pairwise learning to learn the relative bioactivity differences between two compounds within the same assay to circumvent the incompatibility among assays. ActFound further exploits meta-learning to jointly optimize the model from all assays. On six real-world bioactivity datasets, ActFound demonstrates accurate in-domain prediction and strong generalization across assay types and molecular scaffolds. We also demonstrate that ActFound can be used as an accurate alternative to the leading physics-based computational tool FEP+(OPLS4) by achieving comparable performance when using only a few data points for fine-tuning. Our promising results indicate that ActFound could be an effective bioactivity foundation model for compound bioactivity prediction, paving the way for machine-learning-based drug development and discovery. Traditional machine learning methods for drug development struggle with bioactivity prediction due to the limited number of compounds in each assay and assay incompatibilities. Feng et al. developed ActFound, a bioactivity foundation model trained by pairwise learning and meta-learning, that improves the accuracy and generalization of bioactivity prediction.