基于训练的可解释性及其在TCR-表位预测中的应用
Explainability from Training with Applications to TCR-Epitope Prediction
- Tulane University（杜兰大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出基于训练的可解释性（EFT）范式，追踪模型训练中的解释以揭示特征依赖，应用于四种TCR-表位预测模型，并构建TCR-XAI2基准，发现CNN与Transformer学习轨迹不同、α/β链证据冲突及真实与预测结构数据偏好差异。
AI中文摘要：
深度学习模型在人工智能用于科学领域已取得强劲性能，然而其黑箱特性限制了我们对模型如何学习科学任务的理解。现有的可解释性方法在揭示模型如何组织证据以及在学习过程中如何演变方面提供的洞察有限。我们引入了基于训练的可解释性（EFT），这是一种模型无关的范式，在训练过程中追踪模型解释，以解释模型为何依赖特定特征以及如何将这些特征组织为预测性证据。我们将EFT应用于四种最先进的T细胞受体（TCR）-表位预测模型：TCR-SRIM、TULIP、MixTCRpred和NetTCR-2.2，涵盖事后解释和按设计可解释的方法，以及Transformer和CNN架构。为了研究结构信息如何影响模型解释，我们引入了一个基准数据集TCR-XAI2，包含388个独特的实验解析的TCR-表位结构，并辅以使用AlphaFold3、Boltz-2、TCRModel2、tFold-TCR和OpenFold3预测的结构。通过使用EFT和TCR-XAI2，我们证明了：（1）CNN和Transformer模型表现出不同的学习轨迹；（2）TCR α和β链的证据在学习过程中可能发生冲突，限制了联合建模两条链的益处，而MHC信息可缓解这一问题；（3）真实与预测的结构数据在TCR-表位预测中表现出不同的TCR和肽特征偏好，以及模型确定性的不同轨迹。
英文摘要：
Deep learning models have achieved strong performance in artificial intelligence for science, yet their black-box nature limits our understanding of how they learn scientific tasks. Existing methods for interpretability provide limited insight into how models organize evidence and evolve during learning. We introduce explainability from training (EFT), a model-agnostic paradigm that traces model interpretation during training to explain why models rely on specific features and how they organize these features as predictive evidence. We apply EFT to four state-of-the-art T cell receptor (TCR)-epitope prediction models, TCR-SRIM, TULIP, MixTCRpred, and NetTCR-2.2, spanning post-hoc and interpret-by-design approaches as well as transformers and CNNs. To investigate how structural information affects model explanations, we introduce a benchmark, TCR-XAI2, containing 388 unique experimentally resolved TCR-epitope structures, complemented by structures predicted using AlphaFold3, Boltz-2, TCRModel2, tFold-TCR, and OpenFold3. Using EFT with TCR-XAI2, we demonstrate that (1) CNN and transformer models exhibit distinct learning trajectories; (2) TCR $α$ and $β$ evidence can conflict during learning, limiting the benefits of jointly modeling both chains, while MHC information mitigates this; and (3) real versus predicted structural data for TCR-epitope prediction exhibits distinct TCR and peptide feature preferences as well as differing trajectories of model certainty.
