A sparse covarying unit that describes healthy and impaired human gut microbiota development

The human gut microbiome is comprised of many species that all interact in strange ways so that we can be healthy and veer off disease. Given its importance, what is a good way to describe the microbiome?

Before this paper, the idea was to describe by a list of parts—X% of species A, Y% of species B, etc. But this type of description is very incomplete. Imagine seeing a flock of birds or a school of fish and describing the whole system as bird 1, bird 2, bird 3, or fish 1, fish 2, fish 3 and so on. If we did this, we would obviously not capture the collective behavior of the system! This paper created a statistical framework for studying the gut microbiome that was borrowed from the fields of econophysics and protein evolution for inferring the structure of complex systems. By using this approach to analyze the developing microbiome of infants, we showed that there is a very simple structure called an ‘ecogroup’ that is responsible for most of the important bacterial dynamics in the human gut. The ecogroup is only made of 15 bacteria, making it a remarkably low-dimensional representation of a seemingly incomprehensibly complex biological system. Moreover, targeting the ecogroup with a designed food (MDCF-2; see https://www.science.org/doi/10.1126/science.aau4732 for details) helped treat malnutrition in Bangladeshi children. This paper demonstrated both the fundamental and practical value of finding simple descriptions of biological systems created by nature. See Gordon Laboratory (https://gordonlab.wustl.edu/).