Driving Accurate Allergen Prediction with Protein Language Models and Generalization-Focused Evaluation
This work addresses the public health challenge of allergen prediction, offering a computational tool for researchers and practitioners, though it is incremental as it builds on existing protein language models.
The authors tackled the problem of accurately identifying allergen proteins by introducing Applm, a framework based on the xTrimoPGLM protein language model, which outperformed seven state-of-the-art methods in tasks like identifying novel allergens and differentiating allergens among homologs.
Allergens, typically proteins capable of triggering adverse immune responses, represent a significant public health challenge. To accurately identify allergen proteins, we introduce Applm (Allergen Prediction with Protein Language Models), a computational framework that leverages the 100-billion parameter xTrimoPGLM protein language model. We show that Applm consistently outperforms seven state-of-the-art methods in a diverse set of tasks that closely resemble difficult real-world scenarios. These include identifying novel allergens that lack similar examples in the training set, differentiating between allergens and non-allergens among homologs with high sequence similarity, and assessing functional consequences of mutations that create few changes to the protein sequences. Our analysis confirms that xTrimoPGLM, originally trained on one trillion tokens to capture general protein sequence characteristics, is crucial for Applm's performance by detecting important differences among protein sequences. In addition to providing Applm as open-source software, we also provide our carefully curated benchmark datasets to facilitate future research.