蛋白语言模型的解释信号无法准确识别过敏原表位，安全性评估需谨慎。
Residue-Level Attributions in Protein Language Models Do Not Recover Allergen Epitopes
- 构建表位基准测试，量化评估模型解释的准确性
- 多种模型在残基级解释上与已知表位重合度不显著高于随机
- 模型依赖序列物理化学特征而非特异性表位机制，适合研究者验证
深度过敏原分类器在新型食品安全性筛查中日益重要，近期蛋白质语言模型显著提升了蛋白级别过敏原预测能力。然而，其解释是否捕捉生物学有意义信息尚不明确。我们引入基于表位的残基级基准，定量评估蛋白质过敏原模型的归因可信度。在冻结的ESM-2、多任务ESM-2和DeepPlantAllergy模型中，蛋白级分类表现稳健，但分类头的解释信号在AUROC、AUPRC和Precision@k指标下，与标注表位的残基级对齐程度并未显著优于随机。集成梯度识别出对模型重要的残基，但未与已知表位重叠。饱和突变实验进一步表明，分类器可能依赖序列的理化性质与组成特征，而非表位特异性机制。因此，残基级重要性信号不应在未经过定量验证的情况下，被解读为安全筛查或低敏设计的免疫学解释。代码已开源：https://github.com/Jeffateth/XAllergen2.0-paper
原文摘要 · Abstract (English)
Deep allergenicity classifiers are increasingly used in safety screening of novel foods, and recent protein language models have substantially improved protein-level allergenicity prediction. However, whether their explanations capture biologically meaningful information remains unclear. We introduce an epitope-grounded residue-level benchmark for quantitatively evaluating attribution faithfulness in protein allergenicity models. Across frozen ESM-2, multi-task ESM-2, and DeepPlantAllergy, protein-level classification was robust, yet classification-head explanation signals did not significantly exceed random in their residue-level alignment with annotated epitopes across AUROC, AUPRC, and Precision@k. Integrated Gradients identified residues that were functionally important to the model, but not overlapping annotated epitopes. Saturation mutagenesis further suggested classifiers may rely on physicochemical and compositional sequence features rather than epitope-specific mechanisms. Residue-level importance signals should therefore not be interpreted as immunological explanations for safety screening or hypoallergen design without quantitative validation. Code available: https://github.com/Jeffateth/XAllergen2.0-paper
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。