Quantifying Cross-Attention Interaction in Transformers for Interpreting TCR-pMHC Binding

• URL: http://arxiv.org/abs/2507.03197v1
• Date: Thu, 03 Jul 2025 22:18:54 GMT
• Title: Quantifying Cross-Attention Interaction in Transformers for Interpreting TCR-pMHC Binding
• Authors: Jiarui Li, Zixiang Yin, Haley Smith, Zhengming Ding, Samuel J. Landry, Ramgopal R. Mettu,
• Abstract summary: Quantifying Cross-Attention Interaction (QCAI) is a new post-hoc method designed to interpret the cross-attention mechanisms in transformer decoders.<n>QCAI achieves state-of-the-art performance on both interpretability and prediction accuracy under the TCR-XAI benchmark.
• Score: 33.31929328499189
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: CD8+ "killer" T cells and CD4+ "helper" T cells play a central role in the adaptive immune system by recognizing antigens presented by Major Histocompatibility Complex (pMHC) molecules via T Cell Receptors (TCRs). Modeling binding between T cells and the pMHC complex is fundamental to understanding basic mechanisms of human immune response as well as in developing therapies. While transformer-based models such as TULIP have achieved impressive performance in this domain, their black-box nature precludes interpretability and thus limits a deeper mechanistic understanding of T cell response. Most existing post-hoc explainable AI (XAI) methods are confined to encoder-only, co-attention, or model-specific architectures and cannot handle encoder-decoder transformers used in TCR-pMHC modeling. To address this gap, we propose Quantifying Cross-Attention Interaction (QCAI), a new post-hoc method designed to interpret the cross-attention mechanisms in transformer decoders. Quantitative evaluation is a challenge for XAI methods; we have compiled TCR-XAI, a benchmark consisting of 274 experimentally determined TCR-pMHC structures to serve as ground truth for binding. Using these structures we compute physical distances between relevant amino acid residues in the TCR-pMHC interaction region and evaluate how well our method and others estimate the importance of residues in this region across the dataset. We show that QCAI achieves state-of-the-art performance on both interpretability and prediction accuracy under the TCR-XAI benchmark.

Related papers

• Enhanced Prediction of CAR T-Cell Cytotoxicity with Quantum-Kernel Methods [2.1356338340947096]
  Chimeric antigen receptor (CAR) T-cells are T-cells engineered to recognize and kill specific tumor cells.<n>It is possible to modify CAR T-cell phenotype through integrating novel signaling components into the co-stimulatory domains.<n>We show improved learning for specific signaling domains and domain positions, particularly where there was lower information highlighting the potential for quantum computing in data-constrained problems.
  arXiv  Detail & Related papers  (2025-07-30T14:21:32Z)
• Aligned Manifold Property and Topology Point Clouds for Learning Molecular Properties [55.2480439325792]
  This work introduces AMPTCR, a molecular surface representation that combines local quantum-derived scalar fields and custom topological descriptors within an aligned point cloud format.<n>For molecular weight, results confirm that AMPTCR encodes physically meaningful data, with a validation R2 of 0.87.<n>In the bacterial inhibition task, AMPTCR enables both classification and direct regression of E. coli inhibition values.
  arXiv  Detail & Related papers  (2025-07-22T04:35:50Z)
• HistDiST: Histopathological Diffusion-based Stain Transfer [6.197687155055788]
  HistDiST is a Latent Diffusion Model (LDM) based framework for high-fidelity H&E-to-IHC translation.<n>HistDiST significantly outperforms existing methods, achieving a 28% improvement in MRA on evaluations H&E-to-Ki6767.
  arXiv  Detail & Related papers  (2025-05-11T00:19:22Z)
• Enhancing TCR-Peptide Interaction Prediction with Pretrained Language Models and Molecular Representations [0.39945675027960637]
  We present LANTERN, a deep learning framework that combines large-scale protein language models with chemical representations of peptides.<n>Our model demonstrates superior performance, particularly in zero-shot and few-shot learning scenarios.<n>These results highlight the potential of LANTERN to advance TCR-pMHC binding prediction and support the development of personalized immunotherapies.
  arXiv  Detail & Related papers  (2025-04-22T20:22:34Z)
• T-cell receptor specificity landscape revealed through de novo peptide design [2.37499051649312]
  An effective binding between T-cell receptors (TCRs) and pathogen-derived peptides presented on Major Histocompatibility Complexes (MHCs) mediates an immune response.<n>Here, we introduce a computational approach to predict TCR interactions with peptides presented on MHC class I alleles, and to design novel immunogenic peptides for specified TCR-MHC complexes.<n>Our approach provides a platform for immunogenic peptide and neoantigen design, opening new computational paths for T-cell vaccine development against viruses and cancer.
  arXiv  Detail & Related papers  (2025-03-01T22:45:19Z)
• TopoTxR: A topology-guided deep convolutional network for breast parenchyma learning on DCE-MRIs [49.69047720285225]
  We propose a novel topological approach that explicitly extracts multi-scale topological structures to better approximate breast parenchymal structures. We empirically validate emphTopoTxR using the VICTRE phantom breast dataset. Our qualitative and quantitative analyses suggest differential topological behavior of breast tissue in treatment-na"ive imaging.
  arXiv  Detail & Related papers  (2024-11-05T19:35:10Z)
• TCR-GPT: Integrating Autoregressive Model and Reinforcement Learning for T-Cell Receptor Repertoires Generation [6.920411338236452]
  T-cell receptors (TCRs) play a crucial role in the immune system by recognizing and binding to specific antigens presented by infected or cancerous cells. Language models, such as auto-regressive transformers, offer a powerful solution by learning the probability distributions of TCR repertoires. We introduce TCR-GPT, a probabilistic model built on a decoder-only transformer architecture, designed to uncover and replicate sequence patterns in TCR repertoires.
  arXiv  Detail & Related papers  (2024-08-02T10:16:28Z)
• Fast Context-Biasing for CTC and Transducer ASR models with CTC-based Word Spotter [57.64003871384959]
  This work presents a new approach to fast context-biasing with CTC-based Word Spotter. The proposed method matches CTC log-probabilities against a compact context graph to detect potential context-biasing candidates. The results demonstrate a significant acceleration of the context-biasing recognition with a simultaneous improvement in F-score and WER.
  arXiv  Detail & Related papers  (2024-06-11T09:37:52Z)
• Attention-aware contrastive learning for predicting T cell receptor-antigen binding specificity [7.365824008999903]
  It has been verified that only a small fraction of the neoantigens presented by MHC class I molecules on the cell surface can elicit T cells. We propose an attentive-mask contrastive learning model, ATMTCR, for inferring TCR-antigen binding specificity.
  arXiv  Detail & Related papers  (2022-05-17T10:53:32Z)
• Model-based Deep Learning Receiver Design for Rate-Splitting Multiple Access [65.21117658030235]
  This work proposes a novel design for a practical RSMA receiver based on model-based deep learning (MBDL) methods. The MBDL receiver is evaluated in terms of uncoded Symbol Error Rate (SER), throughput performance through Link-Level Simulations (LLS) and average training overhead. Results reveal that the MBDL outperforms by a significant margin the SIC receiver with imperfect CSIR.
  arXiv  Detail & Related papers  (2022-05-02T12:23:55Z)
• Lung Cancer Lesion Detection in Histopathology Images Using Graph-Based Sparse PCA Network [93.22587316229954]
  We propose a graph-based sparse principal component analysis (GS-PCA) network, for automated detection of cancerous lesions on histological lung slides stained by hematoxylin and eosin (H&E) We evaluate the performance of the proposed algorithm on H&E slides obtained from an SVM K-rasG12D lung cancer mouse model using precision/recall rates, F-score, Tanimoto coefficient, and area under the curve (AUC) of the receiver operator characteristic (ROC)
  arXiv  Detail & Related papers  (2021-10-27T19:28:36Z)
• Confidence-guided Lesion Mask-based Simultaneous Synthesis of Anatomic and Molecular MR Images in Patients with Post-treatment Malignant Gliomas [65.64363834322333]
  Confidence Guided SAMR (CG-SAMR) synthesizes data from lesion information to multi-modal anatomic sequences. module guides the synthesis based on confidence measure about the intermediate results. experiments on real clinical data demonstrate that the proposed model can perform better than the state-of-theart synthesis methods.
  arXiv  Detail & Related papers  (2020-08-06T20:20:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.