The landmark US National Institutes of Health-funded Human Microbiome Project (HMP), and its follow-up, the Integrative Human Microbiome Project (iHMP), increased our understanding of how the microbiome affected human health. These projects to map the human microbiome have proved hugely valuable, helping to lay the foundation for further investigation and discovery of core microbial biomarkers. This work has already contributed to disease prevention and prediction, and to precision treatment — and is likely to become even more prominent in future therapies.

As powerful as the HMP was, the data it generated have limitations. Launched in 2007, the HMP was based on a small group of only 242 people, from two US cities: St Louis in Missouri, and Houston in Texas. But my microbiome is not the same as yours: microbiomes are shaped by environmental factors such as diet, lifestyle, genetics and ethnicity. Using a microbiome survey from North America, say, to draw conclusions for an Asian population probably won’t work.

Researchers from both the United States and Europe have established large-scale cohort studies — such as the American Gut Project and Dutch Microbiome Project — to comprehensively capture microbiome profiles in these regions. The results will be immensely valuable to clinical researchers in those locations.

China contains a huge number of populations with diverse diets, cultures and lifestyles. This calls for the country to launch its own large-size cohort study to map the microbiome of its population. But most of the existing human microbiome work in China consists of small to medium-sized studies in specific regions and focusing on the gut microbiome, such as the Guangzhou Nutrition and Health Study and the Guangdong Gut Microbiome Project. Some interesting results are emerging from one nationally representative gut microbiome study, which is a small part of the China Health and Nutrition Survey.

To assemble these studies into something more comprehensive, I co-founded the Westlake Gut Project (WeGut) epidemiology consortium1. All in all, WeGut has microbiome data for 32,000 participants from seven cohorts, covering 17 provinces and megacities across China. This large data set could lay the foundation for future collaboration and pooled analyses. WeGut could also contribute to the setting up of a reference gut microbiome profile for Chinese populations in future.

Beyond the WeGut framework, some existing large longitudinal cohort studies, such as the China Kadoorie Biobank and the Shanghai Women’s and Shanghai Men’s health studies, have started to collect stool samples during follow-up visits for gut microbiome profiling.

This is progress, but it’s not enough.

There are technical limitations. Most of the China-based studies use shallow ‘16S rRNA’ sequencing, meaning there is a lack of large-scale, deeper metagenomic data. Better sequencing would deliver comprehensive profiles of microbial genes and functions.

A more rigorous approach would be to perform microbial metagenomic sequencing, coupled with multiomics measurements of function in the faeces or blood. Biomarkers and fingerprints derived from gut bacteria could then indicate whether or how the bacteria work in the human body. This approach is already used by the iHMP2 and the European Lifelines DEEP cohort3.

Some might argue that there is a lot of hype in the field, and that microbiome-mapping work reports only correlation, not causation. Larger studies, once linked to other investigative efforts, will be able to determine causation — and thus bring microbiome studies into health care faster.

A united effort is needed for the establishment of a ‘Chinese microbiome initiative’. I propose that any such project be built around three key points:

1. Multidisciplinary leadership. Scientists from epidemiology, microbiology, human genetics, nutrition, bioinformatics and other relevant fields are needed to generate a high-quality research protocol with strict data quality control and management.

2. Robust sampling. In addition to the collection of biological samples for measurement of the microbiome, collection of all available clinical, medical imaging, diet and lifestyle data linked to the participants is important to expand the project for wider biomedical research. Any national survey should also aim to record physical activity and physiological data from wearable devices. Such research could also link to medical records. Data must be drawn from a large number of participants, be representative of populations and rely on informed consent from each participant.

3. Transparent data sharing. A pre-defined strategy for sharing and protecting data is necessary to help to facilitate wider use of the data by the research community inside and outside China.

This initiative will not, of course, be free — and would probably require a dedicated funding stream. But the microbiome is the new frontier of medicine, and China is falling behind in terms of understanding the multitudes of bacteria in its multitude of people. Can the country afford not to create a Chinese microbiome project?