COMBO-RATE: An experimentally validated bioinformatic tool to identify promiscuous HLA restrictions

Area of Science:

• Immunology
• Bioinformatics
• Vaccinology

Background:

• Defining HLA restriction of T cell epitopes is crucial for understanding immune responses in infectious diseases, autoimmunity, and vaccine development.
• Existing tools like IEDB RATE infer single HLA restrictions, but T cell epitopes are often presented by multiple HLA alleles (promiscuous restriction), limiting current approaches.
• Promiscuous restriction is a common characteristic of immunodominant epitopes, necessitating methods to identify multi-allelic restriction patterns.

Purpose of the Study:

• To develop COMBO-RATE, an extension of the RATE tool, designed to systematically evaluate combinations of HLA alleles for identifying multi-allelic restriction patterns.
• To address the limitations of single-allele restriction inference by providing a comprehensive framework for defining complete HLA restriction profiles.
• To improve the design of vaccines that require broad HLA coverage across diverse populations.

Main Methods:

• COMBO-RATE was developed as an extension of the existing RATE tool to analyze combinations of HLA alleles.
• The method was applied to three independent datasets covering different antigen systems and epitope discovery strategies.
• Experimental validation involved single-HLA transfected cell lines and antigen presentation assays to confirm COMBO-RATE-inferred restrictions.

Main Results:

• Analysis revealed promiscuous restriction as a near-universal feature of immunodominant epitopes across the tested datasets.
• COMBO-RATE identified multi-allelic restrictions for 32 out of 43 immunodominant CD4+ T cell epitopes from a *B. pertussis* screen, outperforming RATE (24 epitopes).
• The tool uncovered 49 additional allele restrictions, including 30 unique alleles, and experimental validation confirmed the accuracy of COMBO-RATE's predictions.

Conclusions:

• COMBO-RATE offers a robust and scalable framework for defining comprehensive HLA restriction profiles from population response data.
• The ability to identify promiscuous restriction patterns has significant implications for designing vaccines with broad HLA coverage for diverse populations.
• The COMBO-RATE pipeline is accessible as a Python package and a web application, facilitating its use in research and development.