Promised too much? Microbiome research in focus.
“One million microbiomes analyzed in three to five years. The reference database for microbiome-related effects.” These were some bold statements when the project was launched by an international consortium in 2019. But what is the current status?
A central problem of microbiome research remains unsolved to this day: there is a lack of comparable, reliable reference data. Statements such as “if I do this and that, the microbiome behaves in this way” or conversely “because the microbiome has changed in this way, these effects result” are often scientifically weak. Not necessarily because the analysis is poor, but because there is no frame of reference.
Objective and claim of the MMHP
This is exactly where the project comes in. In 2019, an international consortium of technology companies, universities, public research institutions and other partners founded the “Million Microbiomes from Humans Project” with the aim of examining one million microbiomes from ten countries within five years. And this under standardized and comparable conditions. The result should be the world’s largest public reference database of microbiome data, a kind of map of the human microbiome.
The actual status
According to public figures, only around 130,000 samples had been analyzed by the end of 2023. There are no more recent figures. This means that even several years after the start, there is no reliable, consolidated figure that is even close to the million mark. This is not ideal public communication, but it is typical for this type of hybrid project involving industry and academia.
Why is it more difficult to define reference data in the microbiome than in the genome?
In contrast to the human genome, the microbiome is not a stable biological object, but an extremely dynamic ecosystem. While the human genome practically does not change over a person’s lifetime, the human microbiome reacts continuously to nutrition, medication, infections, lifestyle and environmental factors.
In addition, microbiome data is highly context-dependent. Sampling, storage, DNA extraction, sequencing technology and bioinformatics sometimes influence the results more than biological differences between individuals. Two samples from the same individual can differ significantly depending on the methodology used.
Reference data in the microbiome are therefore not fixed standard values, but correspond to statistical reference areas. They describe probabilities, distributions and frequencies, but not ideal states. A deviation from a reference is not automatically pathological, but initially a deviation from a selected methodological framework.
Large-scale projects such as the “Million Microbiomes Human Project” are therefore less an end point than an infrastructure test: they are intended to help understand what is comparable at all before statements can be made about cause, effect or intervention.
One million samples in five years?
130,000 samples correspond to only 13% after 4 years. According to the communicated timeline, 800,000 samples should have already been analyzed at this point.
To understand this discrepancy, it is worth taking a look at how such large-scale projects come about: Public funding is a key factor here. A microbiome research project aiming for 150,000 samples after 5 years has significantly lower chances of attracting attention and funding than a project with a symbolic target figure of one million. Communicative and strategic arguments are particularly important with a higher number of partners and a corresponding amount of funding – even in the academic sector.
In the MMHP, there is no transparently disclosed, central government funding with a clearly named funding amount as in traditional research programs. Instead, it is a consortium of various academic institutions and industrial partners with no public information on an overall funding budget. Government involvement may exist through institutional third-party funding from individual partners, but no coherent public funding amount is officially documented.
The 1 million sample figure was chosen for publicity.
Delays were probably accepted. In addition, the coronavirus pandemic had a negative effect in 2020-2022.
Why the project is still relevant
Nevertheless, the MMHP remains an extremely exciting project in the field of microbiome research. It shows very clearly that in the (micro)biological field, time data can often not be considered linear. While the industrial partners can play to their strengths in throughput and standardization, the collection of clinical and lifestyle-related metadata remains laborious, expensive and fragmented.
The result is large, technically homogeneous data sets with limited biological context depth. This is valuable for method development. For causal statements only to a limited extent.
Furthermore, the MMHP is not a classic study with defined endpoints, but a project to build a data foundation. Such projects have their own dynamics, consisting of a start with grand announcements, but then a slower and more laborious build-up with incremental progress. It is more of a permanently growing reference project than a temporary program. But that doesn’t make it any less attractive.
Another major challenge is standardized sampling, harmonized metadata, comparable sequence quality and central bioinformatics.
These factors significantly determine the quality of a final database and very probably play a central role in the delay to date.
Classification of MMHP microbiome research
Even if it takes another 10 years, the goal of generating a huge public database of microbiome data is fascinating.
However, it remains to be seen to what extent this will be implemented. This is because industry participation involves valid economic interests that potentially stand in the way of full public data release.
The most important questions
What is it about?
An international project in the field of microbiome research relies on highly scaled, standardized sequencing of microbiome samples. The aim is to collect one million samples. Primarily the intestinal and oral microbiome will be analyzed. The data will later be made available via a public platform.
What does that mean?
If the standardization and the quality of the metadata are really as hoped for, a reference will be created that can finally put many typical microbiome statements to the test: What is normal, what is population-dependent, what is methodological?
What impact will this have one day?
A reliable reference from this microbiome research can make diagnostics and monitoring more realistic and practical. Microbiome analyses can finally be matched and provide reliable information. Based on this, more precise nutritional and intervention strategies can be developed and implemented.
Is the project in-time?
No. At the end of 2023, only 13 percent of the samples had been analyzed.
What are the reasons for the delay?
The coronavirus pandemic has played a role in this. But challenges in logistics and qualitative metadata are probably also to blame for the delay.
Has the project failed?
No. In 2023, a major microbiome study was launched in France as part of the MMHP to show how diet, lifestyle and bacterial colonization in the gut interact in the long term. In the “Le French Gut Project”, 100,000 microbiome samples are to be analyzed by 2027.
What do we learn from this?
Microbiome research is complex. Pure sequence analysis is not enough to generate findings from microbiome analyses. Only through careful standardization, method comparisons, transparent limitations and high-quality metadata can a database be created that is suitable as a reference.
Sources:
https://db.cngb.org/mmhp
https://lefrenchgut.fr/the-french-gut-project/
The most important facts at a glance
- A large-scale consortium, the “Million Microbiomes from Humans Project” (MMHP) aims at a standardized microbiome reference with a focus on stool and saliva.
- One million microbiome samples to be analyzed
- The value of such a database stands and falls with standardization, metadata and representativeness, not with the pure number of samples.
- Although large references help with taxonomy and comparability, they are no substitute for causal study designs.
- The project has been severely delayed compared to the original targets.
- The discrepancy between claim and reality is typical for large-scale microbiome research.
- Nevertheless, such a database could deliver new and exciting results. But only if it is publicly accessible.
