AmesFormer: State-of-the-Art Mutagenicity Prediction with Graph Transformers

The Ames mutagenicity test is a gold standard assay for the safety assessment of new chemicals. However, many in silico models rely on challenging-to-interpret ensemble strategies and molecular fingerprint data, which neglects gestalt molecular structure. To improve upon these models, we propose AmesFormer, a graph transformer neural network that shows state-of-the-art performance when paired with our new Ames data set. We briefly review the current state of Ames modeling with a focus on graph neural networks. We then benchmark AmesFormer on a standardized test data set against 22 other Ames models, achieving state of the art (SOTA) performance. We uniquely report the calibration performance of our model and attempt to improve it using temperature scaling. We support our findings with reference to other models from the literature and with developments in machine learning (ML) and graph theory. Overall, we present a high-performance, accessible, and open-source computational model for Ames mutagenicity, with significant potential for regulatory and drug development applications.