hlancpred

hlancpred predicts promiscuous peptide binders for specified non-classical human leukocyte antigen (HLA) alleles (HLA-G*01:01, HLA-G*01:03, HLA-G*01:04, HLA-E*01:01, and HLA-E*01:03) to support immunotherapy and antigen design.

Key Features:

• Artificial intelligence–based prediction: Employs machine learning/artificial intelligence methodologies to predict peptide binders for class-Ib HLA alleles.
• Training data: Models are trained and validated using experimentally verified epitope data from the Immune Epitope Database (IEDB).
• Predictive performance: Validation datasets report AUC values exceeding 0.98 for HLA-G alleles and up to 0.96 and 0.94 for HLA-E*01:01 and HLA-E*01:03, respectively.
• Promiscuous binder identification: Predicts antigenic regions and peptides capable of binding multiple non-classical HLA alleles.
• Viral immunotherapy application: Has been applied to predict non-classical HLA binding peptides within the SARS-CoV-2 spike protein, including Omicron (B.1.1.529).

Scientific Applications:

• Cancer immunotherapy research: Facilitates identification of promiscuous binders to non-classical HLA alleles for therapeutic antigen selection.
• Infectious disease vaccine and immunotherapy development: Assists in predicting peptides from viral proteins, including SARS-CoV-2 variants such as Omicron, that interact with non-classical HLA molecules.
• Antigen design and epitope mapping: Supports selection of antigenic regions that can engage a broad array of non-classical HLA alleles.

Methodology:

Models use artificial intelligence/machine learning and were trained and validated using experimentally verified data from the Immune Epitope Database (IEDB).