Giant test of gut microbiome sequencers:

Polish‑Estonian team shows when Illumina and MGI data can – and cannot – be combined

A large study of over 1,300 paired gut microbiome samples reveals that Illumina and MGI platforms agree on “who is there” in our gut, but differ enough on “what they can do” to require caution in functional analyses.

Why were the two platforms compared?

The study was conducted by researchers from the Małopolska Centre of Biotechnology at Jagiellonian University and the Sano Centre for Computational Medicinein Kraków in collaboration with the University of Tartu in Estonia, using the unique resource of the Estonian Biobank. As more affordable MGI platforms are increasingly used alongside Illumina as the long‑standing standard, an urgent question has emerged: whether data from both technologies can be safely combined in large population‑scale projects.

What was done in practice?

The team analysed 1,351 matched gut microbiome samples, in which exactly the same biological material was sequenced in parallel on Illumina and MGI instruments. This is one of the largest unbiased comparisons of DNA sequencing platforms in human microbiome research conducted to date.

Good news: taxonomic agreement

The two technologies agreed very well on the composition of the microbial community – over 92% of detected species were shared between the platforms. Standard measures of microbial diversity also showed no significant differences, suggesting that existing Illumina and MGI datasets can be combined in taxonomic analyses.

A caveat: differences in microbiome functions

The picture became more complex when the researchers examined the functional potential of the microbiome. In this area they identified systematic differences between the platforms, meaning that uncritical merging of data could lead to misleading conclusions in functional studies.

What does this mean for future research?

The results, published in mSystems together with an accompanying perspective article, provide practical guidance for teams planning large‑scale population health projects using microbiome data. The study highlights that the choice and mix of sequencing platforms is not only a matter of cost and availability, but also a key source of potential bias, especially in analyses of microbiome functions.

Main article: journals.asm.org/doi/10.1128/msystems.01714-25
Perspective article: journals.asm.org/doi/10.1128/msystems.00144-26