On May 9, 2026, the PREANALYTICS project published an article titled “CroCoDeEL: accurate control-free detection of cross-sample contamination in metagenomic data” in Nature Communications.
Cross-contamination between samples, which occurs when DNA is inadvertently exchanged between samples processed simultaneously, is a phenomenon that can skew metagenomic sequencing results of microbial communities. If undetected, this contamination leads to the attribution of microorganisms to a sample that are actually absent, compromising the validity of the results.
Existing methods based on negative controls added to sample plates failed to detect this phenomenon satisfactorily. The article reveals critical contaminations that went unnoticed in several highly cited studies that had previously served as references.
In this article, the authors present a decision-support tool called CroCoDeEL. CroCoDeEL identifies specific contamination patterns in species abundance profiles. It thus addresses this gap by providing highly sensitive detection that works reliably without requiring negative controls or prior knowledge of the order in which samples are processed.
- Lindsay Goulet, Université Paris-Saclay, INRAE
- Florian Plaza Oñate, Université Paris-Saclay, INRAE
- Alexandre Famechon, Université Paris-Saclay, INRAE
- Benoît Quinquis, Université Paris-Saclay, INRAE
- Eugeni Belda, Sorbonne Université, IRD – Inserm
- Edi Prifti, Sorbonne Université, IRD – Inserm
- Emmanuelle Le Chatelier, Université Paris-Saclay, INRAE
- Guillaume Gatreau, Université Paris-Saclay, INRAE