This study, co-authored by researchers from both Bio2Q and Amsterdam UMC, introduces bifidoAnnotator, a bioinformatics tool that enables fine-grained annotation of bifidobacterial enzymes involved in human milk glycan (HMG) metabolism.

Bifidobacteria play a central role in establishing the infant gut microbiome by breaking down complex glycans found in human milk, yet existing annotation tools typically classify glycoside hydrolases only at the family level, limiting their ability to distinguish enzymes with different substrate specificities and metabolic functions. Here, these researchers developed a curated reference database containing more than 22,000 bifidobacterial glycoside hydrolases organized into 108 functional clusters and combined it with a hierarchical annotation framework that identifies specific enzyme variants and associated transport proteins. Benchmarking against established annotation tools showed that bifidoAnnotator achieved superior precision while maintaining perfect recall, and completed analyses an order of magnitude faster than the most accurate existing methods. Applying the tool to infant gut microbiome datasets further revealed species- and strain-level differences in human milk glycan utilization that were not captured by conventional family-level annotations.

By providing accurate, high-resolution functional annotation at scale, this study establishes bifidoAnnotator as a valuable resource for comparative microbiome research, investigations of early-life microbial ecology, and the identification of bifidobacterial strains with probiotic potential.