Back

Culture-enriched metagenomic sequencing reveals within-patient diversity and transmission of vancomycin-resistant Enterococcus faecium

Mills, E.; Evans, K.; Dorazio, A.; Squires, K.; Sundermann, A. J.; Stellfox, M.; Culyba, M.; Shields, R.; Van Tyne, D.

2026-03-16 infectious diseases
10.64898/2026.03.11.26348025 medRxiv
Show abstract

Colonization of the gastrointestinal (GI) tract by vancomycin-resistant Enterococcus faecium (VREfm) often precedes bloodstream infection and serves as a reservoir for onward patient transmission in healthcare settings. Routine clonal isolate-based sequencing often underestimates within-patient diversity, and can miss transmission involving low-abundance and co-colonizing strains. Here we applied culture-enriched metagenomic sequencing to matched GI tract and blood VREfm populations collected [≤]14 days apart from 35 patients with positive VREfm blood cultures collected between 2020 and 2025 at a single hospital. GI populations exhibited greater within-patient diversity than bloodstream populations, including multi-strain colonization in five patients. Among single-strain populations, variant analysis suggested distinct environment-specific pressures between the GI tract and bloodstream environments. To assess transmission using culture-enriched metagenomic sequencing, we compared all 70 VREfm populations against 470 contemporary clinical VREfm isolate genomes collected from the same hospital and identified 19 putative transmission clusters, including six clusters involving multi-strain populations. Together, these results demonstrate how culture-enriched metagenomic sequencing improves resolution for assessing within-patient VREfm diversity and enhances the detection of transmission events that could be missed by clonal isolate-based surveillance. Impact StatementVREfm bloodstream infection is often seeded from bacteria colonizing the gut. The genetic diversity within gut and blood VREfm populations, and the role of this diversity in bacterial transmission, has been difficult to resolve as genomic surveillance typically relies on sequencing a clonal clinical isolate from each patient. Using culture-enriched metagenomic sequencing of matched GI tract and bloodstream VREfm populations from 35 patients at a single hospital, we found that the GI tract reservoir contained VREfm populations with greater strain and variant diversity than populations collected from the bloodstream. By integrating population sequencing with a large collection of VREfm clinical isolate genomes, we further demonstrate that different strains co-colonizing the GI tract of the same patient can reside in multiple putative transmission clusters, revealing potential transmission links that clone-based approaches are likely to miss. These findings demonstrate the potential utility of culture-enriched metagenomic sequencing for higher-resolution hospital surveillance of bacterial transmission. Applying this approach to other bacterial pathogens could improve our ability to detect and interpret transmission involving heterogeneous microbial populations that colonize and infect hospitalized patients. Data SummaryPatient demographic data and clinical characteristics can be found in Table S1 (online Supplementary Material). All sequencing data generated in this study has been deposited in the National Center for Biotechnology Information (NCBI) under BioProject PRJNA901969, with sample accession numbers listed in Table S2. Sequences used to construct the local reference strain database are available at NCBI BioProject PRJNA475751, with accession numbers listed in Table S3. Variants identified in single-strain blood and GI populations are listed in Table S4. Accession numbers for clinical isolate genomes included in transmission analyses are listed in Table S5. RepositoriesSequencing data generated in this study is deposited in the National Center for Biotechnology Information (NCBI) under BioProject PRJNA901969.

Published in Microbial Genomics (predicted rank #5) · training set

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.