# Comparison between metatranscriptomics and viral metagenomics, 16S, and host transcriptomics for comprehensive profiling of the respiratory microbiome and host response

## TL;DR

Metatranscriptomics can efficiently profile the respiratory microbiome and host response in a single test, potentially replacing multiple sequencing methods.

## Contribution

The study demonstrates that metatranscriptomics can replace or complement kingdom-specific methods for microbiome profiling.

## Key findings

- Metatranscriptomics detected RNA viruses and human genes with high concordance and correlated abundance.
- It identified transcriptionally active DNA viruses and revealed distinct bacterial transcriptional profiles.
- Metatranscriptomics reproduced host–microbiome endotypes and showed stronger microbial associations.

## Abstract

Omics-based studies focusing on a single kingdom, such as bacterial 16S gene sequencing, viral metagenomics, and human mRNA sequencing, are commonly used to explore the microbiome and its association with host responses. But combining these approaches is often expensive and time-consuming. Metatranscriptomics provides a snapshot of the entire active microbiome through bulk RNA sequencing in a single test, yet its performance relative to kingdom-specific methods has not been systematically assessed.

We compared metatranscriptomics with three kingdom-specific sequencing approaches in 20 nasopharyngeal aspirates from infants 7 months of age hospitalized for bronchiolitis at the Hospices Civils de Lyon.

Applying specific sequencing depth thresholds (≥1,000 bacterial reads, ≥100,000 human reads, and detection of an internal RNA control), metatranscriptomics showed high detection concordance and correlated abundance for RNA viruses and human coding genes. Metatranscriptomics also detected RNA from both eukaryotic and prokaryotic DNA viruses, suggesting potential for identifying transcriptional activity. For the bacteriome, 82% of genera exceeding 0.5% relative abundance were captured, revealing distinct transcriptional profiles at the species level. Metatranscriptomics reproduced multi-omics-derived host–microbiome endotypes and revealed stronger key microbial associations, particularly for transcriptionally active microorganisms.

These findings indicate that a single metatranscriptomics run can complement or replace kingdom-specific approaches for profiling RNA viruses and the host transcriptome, while also identifying transcriptionally active bacteria and DNA viruses. Low-abundance or latent microorganisms may still require targeted assays. Metatranscriptomics thus provides a cost- and time-efficient strategy for integrated microbiome research and holds promise for clinical applications in acute infections and cases of diagnostic uncertainty.