98%
921
2 minutes
20
Predicting peptide-HLA binding is crucial for advancing immunotherapy; however, current models face several challenges, including peptide length variability, HLA sequence similarity, and a lack of experimentally validated negative data. To address these issues, we present PHLA-SiNet, an efficient pipeline that combines innovative representations with a lightweight architecture. PHLA-SiNet introduces three key components: (1) ESM-Pep, a peptide representation derived from a pre-trained language model (ESM), enabling flexible and training-free embedding of variable-length peptides; (2) IC-HLA, an HLA representation that captures allele-specific discriminative features using information content from binding and non-binding peptides; and (3) SiNet, a Siamese neural network that aligns peptide and HLA embeddings, bringing true binders closer in feature space. Prioritizing sensitivity-essential for reliably identifying candidate binders in early-stage immunotherapy-PHLA-SiNet is rigorously validated across benchmarks, including comparisons of IC-HLA with standard HLA encodings, analysis of SiNet's internal representations, evaluations against 11 leading predictors, and performance assessments across diverse cancer types, rare and unseen HLA types. Through computational validation using molecular docking and analysis of clinical data related to the peptide-based vaccine KIF20A-66, PHLA-SiNet demonstrates strong predictive potential in real-world contexts and highlights HLA-B08:01 as a candidate restriction element for future biological investigation. Overall, PHLA-SiNet offers a generalizable, interpretable, and resource-efficient solution for PHLA binding prediction, with strong promise to accelerate personalized immunotherapy development.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1016/j.compbiomed.2025.111017 | DOI Listing |
Comput Biol Med
September 2025
Computational Biology Research Center (CBRC), Department of Mathematics and Computer Science, Amirkabir University of Technology, Tehran, Iran. Electronic address:
Predicting peptide-HLA binding is crucial for advancing immunotherapy; however, current models face several challenges, including peptide length variability, HLA sequence similarity, and a lack of experimentally validated negative data. To address these issues, we present PHLA-SiNet, an efficient pipeline that combines innovative representations with a lightweight architecture. PHLA-SiNet introduces three key components: (1) ESM-Pep, a peptide representation derived from a pre-trained language model (ESM), enabling flexible and training-free embedding of variable-length peptides; (2) IC-HLA, an HLA representation that captures allele-specific discriminative features using information content from binding and non-binding peptides; and (3) SiNet, a Siamese neural network that aligns peptide and HLA embeddings, bringing true binders closer in feature space.
View Article and Find Full Text PDFCurr Issues Mol Biol
August 2025
Drug Design and Bioinformatics Lab, Faculty of Pharmacy, Medical University of Sofia, Dunav St. 2, 1000 Sofia, Bulgaria.
Tumor immunogenicity depends on the ability of peptides to form stable and specific interactions with both HLA molecules and T-cell receptors (TCRs). While HLA binding is essential, not all HLA-binding peptides elicit T-cell responses. This study investigates the molecular features distinguishing immunogenic T-cell epitopes from non-immunogenic HLA binders.
View Article and Find Full Text PDFImmunooncol Technol
September 2025
Division of Molecular Oncology & Immunology, The Netherlands Cancer Institute, Amsterdam, The Netherlands.
Background: Neoantigen-based immunotherapies rely on computational tools predicting peptide immunogenicity based on properties such as its expression level, binding affinity to human leukocyte antigen (HLA), likelihood of proteasomal cleavage and dissimilarity from wild-type peptide. However, current datasets are scarce and limited to highly mutated tumor types such as melanoma and lung cancer, leaving uncertainty about the value of these properties in other tumor types.
Materials And Methods: To investigate this, we retrospectively analyzed the properties of immunogenic neoantigens identified in CD8 T-cell recognition screens of predicted neoantigens in tumor-infiltrating lymphocytes (TILs) from 12 melanoma patients and peripheral blood mononuclear cells (PBMCs) from 14 patients with mesothelioma, triple-negative breast cancer or urothelial cancer.
Nat Genet
August 2025
Department of Dermatology, Stanford University, Stanford, CA, USA.
Human leukocyte antigens (HLAs) are encoded by the most polymorphic genes in the human genome. HLA class I alleles control antigen presentation for T cell recognition, which is pivotal for autoimmunity, infectious diseases and cancer. Current knowledge of HLA-bound peptides is limited, skewed and falls short of population-wide HLA binding profiles for high-value targets.
View Article and Find Full Text PDFJ Phys Chem B
July 2025
Institute of Quantitative Biology, School of Physics, and College of Life Sciences, Zhejiang University, 310027 Hangzhou, China.
Human CD8+ cytotoxic T lymphocytes (CTLs) induce melanoma regression by utilizing T-cell receptors (TCRs) to recognize tumor-specific epitopes. One such epitope, gp100 (YLEPGPVTV), is presented by human leukocyte antigen (HLA)-A*02:01 molecules. Computational approaches, particularly the predictions on peptide-HLA (pHLA) and TCR-pHLA binding affinities, are complementary to experimental approaches in guiding further optimization and design of the gp100 peptide (mimotopes) for enhanced antitumor responses.
View Article and Find Full Text PDF