Metagenomic Investigation of Plasma in Individuals with ME/CFS Highlights the Importance of Technical Controls to Elucidate Contamination and Batch Effects.

Evidence Atlas

Miller, Ruth R, Uyaguari-Diaz, Miguel, McCabe, Mark N et al. · PloS one · 2016 · DOI

Quick Summary

Researchers used advanced genetic sequencing to search for viruses and bacteria in the blood of people with ME/CFS, comparing them to people with other conditions and healthy individuals. They found very little microbial material in anyone's blood, and the small differences they observed were likely caused by contamination during sample handling rather than actual differences between groups. This study teaches an important lesson: when looking for germs in blood samples, scientists must be extremely careful about how they handle samples and must use proper controls to avoid false conclusions.

Why It Matters

This study is important because it demonstrates a critical methodological principle: many previous studies claiming to find specific microbes in ME/CFS blood may have produced false positive results due to contamination and batch effects. Understanding these technical limitations helps researchers design better future studies and prevents the field from pursuing false leads about infectious causes. This work ultimately protects ME/CFS patients by ensuring that future research is built on solid scientific foundations.

Observed Findings

Majority of sequencing reads were removed during host subtraction, leaving very low microbial RNA content in plasma from all study groups.
Bacterial taxa including Proteobacteria, Firmicutes, Actinobacteria, and Bacteroides were detected in both patient plasma samples and plasma-free negative control samples.
Sample batch and processing contamination effects outweighed differences between ME/CFS, chronic Lyme syndrome, SLE, and healthy control groups.
Considerable overlap existed between bacterial sequences in study samples and technical controls, with many organisms previously reported in disease literature.
Few observed differences in bacterial or viral RNA abundance between groups were likely attributable to contamination and batch effects rather than disease-related biology.

Inferred Conclusions

Negative controls and batch effect analysis are essential in metagenomic studies and have been underutilized in prior research.
Previously published associations between specific microorganisms and ME/CFS may reflect methodological artifacts rather than true biological findings.
Plasma metagenomic analysis has significant technical limitations that must be overcome before reliable conclusions about microbial content can be drawn.
Current protocols for plasma sample processing introduce contamination that confounds interpretation of microbial content differences between disease and control groups.

What This Study Does Not Prove

This study does not prove that ME/CFS has no infectious cause—it only shows that plasma metagenomic sequencing as currently performed may not be a reliable method for detecting microbes in ME/CFS. The study also does not address whether microbes in other tissues (not blood) might be involved in ME/CFS pathology. The findings do not eliminate infection as a potential contributor, but rather highlight that better technical approaches are needed to investigate it.

Metadata

DOI: 10.1371/journal.pone.0165691
PMID: 27806082
Review status: Editor reviewed
Evidence level: Early hypothesis, preprint, editorial, or weak support
Last updated: 12 April 2026

About the PEM badge: “PEM required” means post-exertional malaise was an explicit required diagnostic criterion for participant inclusion in this study — not that PEM was studied, observed, or discussed. Studies using criteria that do not require PEM (e.g. Fukuda, Oxford) are tagged “PEM not required”. How the atlas works →

Evidence Atlas

Metagenomic Investigation of Plasma in Individuals with ME/CFS Highlights the Importance of Technical Controls to Elucidate Contamination and Batch Effects.

Quick Summary

Why It Matters

Observed Findings

Inferred Conclusions

What This Study Does Not Prove

Tags

Metadata

Explore further

Remaining Questions