dbPepNeo: a manually curated database for human tumor neoantigen peptides

Area of Science:

• Oncology
• Immunology
• Bioinformatics

Background:

• Neoantigens are critical for tumor rejection and personalized cancer vaccines.
• Experimental validation of neoantigens is essential but challenging due to their scarcity and scattered data.
• Existing computational predictions require robust experimental validation for clinical application.

Purpose of the Study:

• To construct a comprehensive database of experimentally validated human neoantigen peptides.
• To provide a centralized resource for neoantigen research in cancer immunology and immunotherapy.
• To facilitate the screening and confirmation of potential neoantigens for clinical use.

Main Methods:

• Compilation of human leukocyte antigen class I (HLA-I) binding neoantigen peptides.
• Inclusion of peptides validated by mass spectrometry (MS) or immunoassay in human tumors.
• Classification of neoantigens into high, medium, and low confidence levels based on validation methods.

Main Results:

• Construction of dbPepNeo, a database of experimentally validated neoantigens.
• Categorization of 295 high, 247 medium, and 794 low confidence neoantigens.
• Demonstration of three distinct applications for the dbPepNeo database.

Conclusions:

• dbPepNeo serves as a valuable resource for advancing cancer immunotherapy research.
• The database supports the optimization of neoantigen prediction pipelines and T-cell receptor (TCR) recognition studies.
• dbPepNeo promotes the screening and confirmation of potential neoantigens, accelerating their clinical application.