A pioneering multi-modal framework with TCR-peptide-HLA sequence and structure features incorporating an attention mechanism designed to accurately identify tumor antigens with immunogenic properties, which presents a brand-new and promising approach for cancer immunotherapies that target tumor antigens.
The ImmunoFoundation Model (IFM), a multimodal deep learning system that integrates not only peptide sequences, 3D molecular structures, and biochemical properties but also TCR-MHC-peptide to achieve superior immunogenicity prediction and enable peptide optimization for therapeutic applications is developed.
Smita Krishnaswamy, J. Rocha, Hiren Madhu et al.· Journal of Immunology· 0 citations
This work is the first to use a multi-modal foundation model for neoantigen vaccines, overcoming the tumor-type-specific constraints of conventional approaches and facilitating joint modelling of immunogenicity and clinical efficacy, thereby offering an AI decision engine for precision cancer vaccine design that is applicable to all cancer types.
Gang Liu, Jia Wang, Jia Zhu· Global Health Care· 1 citation
The current role of AI is summarized across the peptide cancer vaccine development pipeline, from neoantigen discovery and epitope prioritization to prediction of peptide–HLA binding, antigen presentation, and T-cell receptor recognition, and the application of modern computational frameworks.
P. Brlek, J. Kolić, L. Bulić et al.· Frontiers in Genetics· 0 citations
AAMFM, an Antigen-specific Antibody Multimodal Foundation Model that learns unified representations of antibody sequences and structures conditioned on antigen context, achieves state-of-the-art performance in functional antibody design, revealing its potential for antigen-specific antibody engineering.
Xiaoliang Shi, Zichen Wang, Runze Ma et al.· arXiv.org· 0 citations
Breast cancer, particularly hormone receptor-positive (HR+) and triple-negative (TNBC) subtypes, is often immunologically "cold," limiting immunotherapy efficacy. Neoantigen-based vaccines hold promise but face challenges from inaccurate identification and overreliance on HLA binding affinity, which poorly captures true immunogenicity. Most approaches focus on patient-specific (private) neoantigens, hindering scalability. In contrast, recurrent driver mutations in ESR1 and PIK3CA produce shared (public) neoantigens suitable for off-the-shelf vaccines, yet systematic discovery and design frameworks for breast cancer are lacking. We developed NeoGen-BC, a synergistic framework advancing from neoantigen prediction to rational design. It integrates protein language model (PLM) embeddings with Retrieval-Augmented Generation (RAG) and a multi-scale MCNN-BiLSTM classifier to capture immunogenic features beyond peptide-HLA binding. A sliding-window strategy maps epitopes from driver mutations, while a controlled ProtGPT2 generative module explores immunogenic peptide space. Candidates are refined via biophysical filtering, structural validation, and immunogenicity screening. NeoGen-BC achieved an AUC of 0.9053 on an independent test set, outperforming other machine learning models. It identified immunogenic peptides difficult to assess by MHC-II binding tools like NetMHCIIpan 4.3 and accurately detected validated MHC class II-restricted shared neoantigens from ESR1 (Y537S, D538G) and PIK3CA (H1047R) mutations, often missed by conventional predictors. De novo generated peptides matched experimentally validated ESR1 neoantigens in physicochemical and structural properties, showing favorable MHC-II binding and immunogenicity. NeoGen-BC provides a computational foundation for next-generation "tunable antigen" vaccine design by prioritizing immunogenicity alongside peptide-MHC binding characteristics. The framework enables scalable identification of candidate peptide vaccines targeting shared breast cancer vulnerabilities and may complement emerging therapeutic strategies, including oral SERD-based approaches aimed at enhancing immune modulation and advancing precision immunotherapy.
Van-The Le, Juan Peter Timothy Yuune, Jiun-I Lai et al.· Biochemical and Biophysical...· 0 citations
Accurate identification of interactions between T-cell receptors (TCRs) and antigenic peptides presented by major histocompatibility complex (MHC) molecules is essential for advancing precision immunotherapy. However, existing approaches often exhibit limited generalization to unseen peptides and struggle to capture the complex interaction patterns underlying immune recognition. Here, we present TCR-IFNet, a biologically informed deep learning framework for interpretable TCR-peptide interaction prediction. The model integrates global contextual representations from protein language models with local motif refinement via a gated convolutional module. To model cross-sequence dependencies, we introduce a Fast Kolmogorov-Arnold Network (FastKAN)-based cross-attention mechanism for nonlinear interaction modeling, together with a bilinear attention network to aggregate residue-level features into compact interface representations. Evaluation across multiple settings indicates that TCR-IFNet achieves competitive performance compared with existing methods, with higher AUPRC observed on both antigen-specific and healthy-sourced datasets, as well as improved results on independent test sets. The model also shows consistent generalization to unseen peptides under different negative sampling strategies. In addition, TCR-IFNet provides biologically meaningful interpretability by identifying key residue-level interaction patterns consistent with structural binding interfaces. Collectively, these findings demonstrate that TCR-IFNet provides a robust and generalizable computational framework for characterizing TCR-peptide interactions.
Wen-Yu Xi, Ruheng Wang, Xiu-Cai Ye et al.· International Journal of Bio...· 0 citations
A weeklong summer workshop brought higher education faculty to campus to explore how AI and machine learning materials can be adapted for their classrooms.
