Updated project metadata. We proposed a complete workflow for neoantigen identification and validation based solely on MS immunopeptidomics.We first developed a new Neo DiscoveryTM immunopeptidome enrichment Kit, which can be used for HLA peptide purification from a small amount of biopsied tissue as low as 18mg. To achieve high sensitivity with such a low amount of sample, we leveraged the power of deep learning and a massive amount of MS-based immunopeptidomics data to build DeepNovo-HLA, a de novo sequencing model specialized for HLA peptides. We also developed DeepImmu, a personalized model for immunogenicity prediction based on the central tolerance of T cells, i.e. the positive and negative selection of T cells in an individual patient.