Library strategies differentially shape microbial, functional, and host signals in clinical metagenomic sequencing
9 affiliations
- 1National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases, Zhongshan School of Medicine, Shenzhen Campus of Sun Yat-sen University, Sun Yat-sen University, Shenzhen, China
- 2State Key Laboratory for Biocontrol, Zhongshan School of Medicine, Shenzhen Campus of Sun Yat-sen University, Sun Yat-sen University, Shenzhen, China
- 3Department of Laboratory Medicine, Zhongnan Hospital of Wuhan University, Wuhan, China
- 4Hubei Provincial Clinical Research Center for Molecular Diagnostics, Wuhan, China
- 5Department of Electrical Engineering, City University of Hong Kong, Hong Kong, China
- 6National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, China
- 7Sydney Institute for Infectious Diseases, The University of Sydney, Sydney, New South Wales, Australia
- 8Centre for Virus Research, Westmead Institute for Medical Research, Sydney, New South Wales, Australia
- 9Institute of Parasitic Disease Control and Prevention, Center for Disease Control and Prevention of Guangdong Province, Guangzhou, China
Abstract
Metagenomic next-generation sequencing (mNGS) is increasingly used in infectious disease diagnostics, yet how library preparation shapes the microbial, functional, and host signals recovered from clinical samples remains poorly defined. Here, we performed a within-sample parallel comparison of three mNGS library preparation strategies—DNA-based libraries (DNAlib), RNA-based libraries (RNAlib), and total nucleic acid–based libraries (TNAlib)—across a diverse range of clinical specimens spanning five sample types. Using a curated clinical infectome as a benchmark, we show that library strategies are not interchangeable but capture distinct biological dimensions of the same specimen. RNAlib provided the most comprehensive standalone recovery of the clinical infectome, with improved detection of RNA viruses and cellular pathogens, enhanced resolution of resistance and virulence signals, and preservation of infection-associated host immune signatures. DNAlib showed stronger baseline recovery of DNA viruses and broader host genome coverage, whereas the TNAlib workflow evaluated here largely behaved as an intermediate strategy rather than a consistent improvement over dedicated DNA- or RNA-based workflows. Together, these results establish that the library preparation protocol is a major determinant of how clinical mNGS data should be interpreted and provide a framework for selecting sequencing strategies according to specific diagnostic and biological questions.
IMPORTANCE: Metagenomic sequencing is increasingly used in infectious disease research and clinical diagnostics, but different library preparation strategies may recover fundamentally different biological signals from the same sample. These signals include not only pathogens but also background microbes, microbial functional activity, and host immune-response patterns. Here, we systematically compared DNA-, RNA-, and total nucleic acid–based metagenomic sequencing libraries using the same clinical samples processed in parallel. We found that the three strategies did not provide equivalent information. RNA-based sequencing generated the most informative single-library view of infection, particularly for RNA viruses, cellular pathogens, functional microbial signals, and host immune-response patterns. DNA-based sequencing was more effective for DNA virus and host genome recovery, whereas the total nucleic acid sequencing workflow evaluated here generally behaved as an intermediate strategy. These findings show that library preparation can substantially influence the interpretation of metagenomic data.
Related
- Clinical severity score-guided metagenomic analysis identifying gut microbial taxonomic and functional markers for severe hepatitis E progression
- Acarbose differentially affects gut microbial metabolism of maltose and sucrose, distinctly modulating host metabolism
- Metagenomic and metabolomic analysis of a bryozoan-associated microbial complex from the Hanjiang River, China
- BioSIFTR: functional inference from shallow shotgun sequencing using biome-specific genome catalogues benchmarked in gut microbiomes
- Microbial communication networks governing the ecosystem dynamics of Salmonella in poultry environments: ecological interactions, colonization resistance, and emerging control strategies
BibTeX
@article{luo2026library,
title = {{Library strategies differentially shape microbial, functional, and host signals in clinical metagenomic sequencing}},
author = {Gengyan Luo and Jin Zhao and Lin-lin Xie and Yirong Li and Yang-zhi Rao and Yanni Sun and Xin Lu and John-Sebastian Eden and Liangjun Chen and Mang Shi},
journal = {mSystems},
year = {2026},
month = sep,
doi = {10.1128/msystems.00897-26},
url = {https://doi.org/10.1128/msystems.00897-26},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Library strategies differentially shape microbial, functional, and host signals in clinical metagenomic sequencing
AU - Gengyan Luo
AU - Jin Zhao
AU - Lin-lin Xie
AU - Yirong Li
AU - Yang-zhi Rao
AU - Yanni Sun
AU - Xin Lu
AU - John-Sebastian Eden
AU - Liangjun Chen
AU - Mang Shi
PY - 2026
DA - 2026/09/14
JO - mSystems
DO - 10.1128/msystems.00897-26
UR - https://doi.org/10.1128/msystems.00897-26
AB - Metagenomic next-generation sequencing (mNGS) is increasingly used in infectious disease diagnostics, yet how library preparation shapes the microbial, functional, and host signals recovered from clinical samples remains poorly defined. Here, we performed a within-sample parallel comparison of three mNGS library preparation strategies—DNA-based libraries (DNAlib), RNA-based libraries (RNAlib), and total nucleic acid–based libraries (TNAlib)—across a diverse range of clinical specimens spanning five sample types. Using a curated clinical infectome as a benchmark, we show that library strategies are not interchangeable but capture distinct biological dimensions of the same specimen. RNAlib provided the most comprehensive standalone recovery of the clinical infectome, with improved detection of RNA viruses and cellular pathogens, enhanced resolution of resistance and virulence signals, and preservation of infection-associated host immune signatures. DNAlib showed stronger baseline recovery of DNA viruses and broader host genome coverage, whereas the TNAlib workflow evaluated here largely behaved as an intermediate strategy rather than a consistent improvement over dedicated DNA- or RNA-based workflows. Together, these results establish that the library preparation protocol is a major determinant of how clinical mNGS data should be interpreted and provide a framework for selecting sequencing strategies according to specific diagnostic and biological questions. IMPORTANCE: Metagenomic sequencing is increasingly used in infectious disease research and clinical diagnostics, but different library preparation strategies may recover fundamentally different biological signals from the same sample. These signals include not only pathogens but also background microbes, microbial functional activity, and host immune-response patterns. Here, we systematically compared DNA-, RNA-, and total nucleic acid–based metagenomic sequencing libraries using the same clinical samples processed in parallel. We found that the three strategies did not provide equivalent information. RNA-based sequencing generated the most informative single-library view of infection, particularly for RNA viruses, cellular pathogens, functional microbial signals, and host immune-response patterns. DNA-based sequencing was more effective for DNA virus and host genome recovery, whereas the total nucleic acid sequencing workflow evaluated here generally behaved as an intermediate strategy. These findings show that library preparation can substantially influence the interpretation of metagenomic data.
SN - 2379-5077
ER -