Multi-omics and bioinformatics pipelines for personalized neoantigen discovery: current frameworks, challenges, and clinical translation

Personalized neoantigen vaccines promise precise cancer immunotherapy, yet identifying immunogenic epitopes remains challenging. This review evaluates bioinformatics and multi-omics pipelines across a four-tier framework: variant-to-candidate prediction, HLA binding and presentation, immunopeptidomics, and proteogenomic integration. We examine MHC-II modeling, long-read sequencing, and TCR-aware architectures while addressing biological bottlenecks, such as tumor heterogeneity and immune evasion, to accelerate the development and clinical translation of robust, personalized cancer immunotherapy.