A roadmap for T cell receptor-peptide-bound major histocompatibility complex binding prediction by machine learning:...

Area of Science:

• Immunology
• Computational Biology
• Bioinformatics

Background:

• Cytotoxic T lymphocytes (CTLs) are vital for fighting cancer and infections.
• T cell receptors (TCRs) mediate CTL activation by recognizing peptide-bound MHC (pMHC).
• Identifying antigen-specific CTLs and their TCRs is key for T-cell based therapies but is experimentally demanding.

Purpose of the Study:

• To review the biological basis of TCR-pMHC binding.
• To compare state-of-the-art machine learning algorithms for TCR-pMHC prediction.
• To propose a roadmap for advancing TCR-pMHC prediction accuracy.

Main Methods:

• Literature review of TCR-pMHC binding mechanisms.
• Comparative analysis of machine learning algorithms for TCR-pMHC prediction.
• Identification of discrepancies in current machine learning methods for specific diseases.

Main Results:

• Machine learning for TCR-pMHC prediction is gaining traction with single-cell technology advancements.
• Existing machine learning methods show limitations under specific disease conditions.
• A roadmap is proposed to enhance TCR-pMHC prediction by improving data, models, and application contexts.

Conclusions:

• Accurate TCR-pMHC prediction is essential for developing effective T-cell based vaccines and therapies.
• Addressing data quality, encoding, model training, and application context is crucial for progress.
• This review provides a framework to guide future research in TCR-pMHC prediction.