A new study from Ran Blekhman PhD’s lab at the University of Chicago and the University of Minnesota introduces an innovative solution to this challenge. Published on January 22, 2025, in Cell, the research presents the Human Microbiome Compendium the largest publicly available dataset of its kind. This initiative standardized the processing of over 160,000 gut microbiome samples from 68 countries marking the first effort of this scale.

Our goal is to compile existing studies on diverse microbiomes worldwide spanning different ages and health conditions into a single standardized dataset said co-lead author Samantha Graham a graduate student at the University of Minnesota. This allows us to uncover patterns that were previously impossible to detect.

16S amplicon sequencing a technique for identifying microorganisms based on a specific genomic region has been a cornerstone of microbiology due to its affordability and simplicity. Although the core steps of 16S sequencing are consistent across studies variations in data collection and processing between labs make it challenging to merge findings from different projects.

In this new study rather than relying on summary tables from various publications researchers undertook the ambitious task of uniformly re-processing raw data from 168,464 samples.

At first glance this may seem like a straightforward albeit time consuming task of revisiting hundreds of individual studies. However the researchers discovered that some of the biggest technical challenges arose from the decisions made by the original study authors.

Each researcher provides sample data in their own unique way. Sometimes it's missing entirely sometimes it's extremely detailed but most often it's in vastly different formats said study author Richard Abdill, PhD, lab manager for Blekhman’s group.

Once the team organized this vast dataset the sheer volume of samples enabled them to explore new research questions.

Some regions like the U.S., have been extensively sampled while others such as Northern Africa and Western Asia have limited publicly available data. The researchers showed that adding even a few samples from underrepresented areas significantly enhances microbial discovery.

They also highlighted issues in generalizing microbiome patterns globally as key microbes linked to conditions like obesity and IBD show different abundances outside Europe and North America.

The team hopes the Human Microbiome Compendium will aid future research particularly in machine learning applications and provide insight into the global disparities in microbiome studies.

Source: https://biologicalsciences.uchicago.edu/news/gut-microbiome-data-underscores-gaps-global-representation