bacNeo: A Computational Toolkit for Identifying Bacteria‐Derived Neoantigens in Human Cancers
iMetaMed · 2026-06 · 2 (2)
DOI: 10.1002/imm3.70040 ↗Abstract
ABSTRACT Intratumor microbiome can rewire the tumor microenvironment (TME) by presenting bacterial neoantigens, while integrated in silico tools for identifying bacterial neoantigens remain lacking. We developed bacterial neoantigen ( bacNeo ), a multi‐omics‐based computational framework that contains three modules to classify bacterial components ( BACC ), type bacteria‐bound HLA alleles ( BACH ), and prioritize bacterial peptides as neoantigens ( BACP ) in human cancers. The input options of bacNeo are diverse: we support metagenomic or metatranscriptomic data, as well as sequencing data in different resolutions, including bulk and single‐cell sequencing; aside from sequencing data, mass spectrometry (proteome) data are also recommended for peptide discovery. The neoantigen potential of these peptides is quantified by score for peptide antigenicity recognition & kinetics (SPARK). Artificial neural networks and gradient boosted decision trees enhance the predictive accuracy of bacNeo , which has been validated in bacterial peptide‐HLA experimental pairs. In conclusion, bacNeo can enable personalized neoantigen discovery to accelerate therapeutic strategies across diverse cancer types.
EUVIMED