Science: MIT Hu Backbone and others use physics tools to help understand complex ecosystems such as the gut microbiota

The human gut is not only a place for digestion and absorption of nutrients, but also a paradise for microorganisms, where hundreds of millions of microorganisms grow wildly, and they interact with each other in complex ways, thus affecting human obesity, enteritis, autoimmune diseases, responses to cancer treatment drugs, and even human lifespan

But now, a new study could help microbiologists achieve this impossible task

The paper, titled Emergent phases of ecological diversity and dynamics mapped in microcosms, was published in the journal Science on October 6, 2022, with the first author Dr.

Dr.

The study was also highly praised by reviewers as a significant contribution to ecology, particularly in the field of microbial community ecology, and as an excellent experimental validation

Ecology strives to understand the diverse species and complex dynamic behaviors in natural ecosystems, yet scientists have long lacked a unified framework

Scientists such as Hu and Jeff Gore of MIT's Department of Physics, combined theory and microbial community experiments, have shown that it is possible to predict the behavior

1.

A large number of species coexist and interact in nature, forming complex ecological communities

Experiments in a laboratory-controlled environment can effectively avoid random interference with environmental factors, so as to study community intrinsic properties, such as the effects of interspecific interactions on

For a large number of species coexisting in laboratory communities, since it is not possible to measure all biological details (interspecific interactions, growth rate, environmental capacity, etc.

Robert May and other pioneering theorists explored predicting the behavior

In this work, scientists such as Hu Jingliang and Jeff Gore have attempted to control the number of species and the intensity of interspecies interactions in theory and experiments, and to reveal the relationship between

2.

With the help of the generalized Lotka-Volterra model, the authors studied changes in

In the simulation, the authors randomly generate interspecies interaction matrices according to a certain statistical distribution, and prove that the kinetic phase and phase transition do not change with the change

The results of the theoretical analysis show that only the three coarse-grained parameters of species number, first-order moment and second-order moment of interspecies interaction distribution can predict the diversity and stability
of complex communities.
The authors further verify that kinetic phase maps are robust under different model assumptions (e.
g.
, considering predator-predator-predator in ecological networks, mutually beneficial symbiosis, competitive relationships, etc.
) and exhibit the same kinetic phase and phase transition sequence
.
The authors even obtained qualitatively identical phase diagrams
in other types of community dynamics models, such as pH-based interspecies interaction models.
These results demonstrate the universality
of community dynamics and phase maps of biodiversity.

Figure 1.
The theoretical prediction of the number of species and the intensity of interspecies interactions shaped the phase map
of the community.
As the number of species and the intensity of interspecies interactions increased, phase transitions occurred between the three emerging kinetic phases of the community, from the stable coexistence phase of all species to the stable coexistence phase of some species, and finally to the continuous oscillation phase
of species numbers over time.

3.
The dynamic behavior of microbial communities from steady state to oscillation

To experimentally verify the kinetic oscillations predicted by the theory, the authors experimentally used 48 different bacteria isolated from soil to form different microbial communities with different combinations
of bacteria.
By varying the concentration of nutrients (glucose and urea) in the culture solution, the authors found that the intensity of interactions between bacteria increased significantly with increased nutrient concentrations
.
Exactly as the theory predicts, a systematic increase in the number of species and the intensity of interspecies interactions in a microbial community in an experiment will cause the composition of species within the community to continue to oscillate
over time.
This continuous population oscillation is reflected in both the violent oscillation of total biomass over time and the violent oscillation
of the proportion of different species over time.
The total organism of the microbial community (Biomass OD) and the proportion of different species (16s sequencing results) exhibit highly consistent results, with both properties of the same community either reaching a steady state at the same time or oscillating
at the same time.

Figure 2.
Microbial community experiments demonstrate theoretical predictions that as the number of species and the intensity of interspecies interactions increase, more and more microbial communities exhibit continuous oscillations
over time.

4.
Phase mapping of experimental microbial community dynamics and biodiversity

By analyzing the biodiversity and stability of microbial communities under different species numbers and interaction intensities in the experiments, the authors experimentally verified the phase maps and phase transitions emerging in ecosystems, and the results were highly consistent
with the theory.
Specifically, experimental ecosystems exhibit the stable coexistence of all species in the parameter space of small number of species and weak interspecies interactions, and the first second-order phase transition will first occur when the number of species is low and the interaction between species is increasing, and some species (species extinction) are lost and transformed into stable coexistence of some species, followed by the second phase transition, and the community loses stability and continues to oscillate
.

In summary, as ecosystems become more complex, communities lose species diversity before they begin to lose their dynamic stability
.
It is worth noting that species survival in ecosystems (the number of surviving species compared to the total number of species) declined rapidly in Phase II (the stable coexistence phase of some species), but no longer decreased significantly in phase III and reached relative stability
.
The next chapter of this article will explain how kinetic oscillations hinder the rapid loss
of species diversity.

Figure 3.
Phase diagram of experimental microbial communities: In the parameter space where the number of species and the intensity of interspecies interactions are coordinated, microbial communities exhibit three different kinetic phases
.
As the number of species and the intensity of interspecies interactions gradually increases, communities undergo two phase transitions, and communities lose some species diversity before losing kinetic stability and begin to oscillate
continuously.

5.
There is positive feedback between population oscillation and species diversity

The theory predicts that as species survival rates in ecosystems (the number of surviving species over the total number of species) declines rapidly first and then into a gentle interval, both survival rates no longer decline rapidly but tend to plateau
.
More interestingly, the calculations show that under the same conditions, oscillating communities always exhibit higher biodiversity
than stable communities.

The authors analyzed the experimental data and found that there was a strong positive feedback between community oscillations and high species diversity that were highly consistent with theoretical predictions
.
The protective effect of kinetic oscillations on species diversity can be understood as the concussion of effective ecological niches over time provides the possibility for the
survival of more species.
Imagine that one group of species and another group of species have strong competition inhibition and cannot coexist, then if the two groups of species maintain a certain phase difference over time, they can allow both groups to grow in different time intervals and achieve "coexistence"
in the sense of time average.

Figure 4.
Theoretical and experimental results consistently show that oscillating communities exhibit higher biodiversity
than stable communities.

Conclusion and outlook

This proposes an effective framework that brings together two of the most famous results of theoretical ecology: on the one hand, Robert May argues that the increase in the complexity of ecological networks necessarily leads to its instability; Peter Chesson, on the other hand, demonstrated that ecosystems can sustain species diversity
over time.

The relationship between biodiversity and community stability has been debated in the field of ecology, and the main reason for this controversy is that the complex dynamics exhibited by natural ecosystems may be caused both by random shocks of the environment and by the intrinsic properties of ecological networks (complex interspecies interaction networks
).
This experimental system effectively controls environmental noise, proving the conclusion of theoretical prediction: only two coarse-grained parameters, namely the number of species and the intensity of interspecies interactions, can effectively describe the dynamic behavior of complex ecosystems
.
These predictions and theoretical frameworks are robust to biological detail, and similar ecodynamic phases can be obtained using resource-consumer models or pH models
.
Therefore, the phase maps of biodiversity and community dynamics proposed in this study may be widely applicable in more ecosystems
.
Future work should attempt to explore whether the kinetic phase diagrams proposed by the study are universally applicable to complex ecological communities of various life forms at various spatiotemporal scales
.

This work may be of interest
to scientists in different fields.
First, the stability and diversity of the microbial community is critical to the function and health of different microbiomes (e.
g.
, intestinal flora and soil flora
).
In addition, several types of ecodynamic models used in this study are widely used in the study of many other ecosystems, so the ecodynamic phase diagram presented here may be universal to other ecological communities
.
Finally, the study proposes a theoretical framework inspired by statistical physics that extracts a small amount of coarse-grained control variables from high-dimensional ecological networks, which may be generalized to the study of
other complex systems.

Jeff Gore (left), Hu Backbone (right)

Jeff Gore is a professor in the Department of Physics at the Massachusetts Institute of Technology (MIT) and serves as the founder and director
of the MIT Physics of Living System Center.
Jeff Gore earned his bachelor's degree in physics, mathematics, electronics, and finance at MIT before joining the UC Berkeley Professor Carlos Bustamante lab to complete his doctoral thesis
in single-molecule physics.
During his Ph.
D.
, he studied the tensile, torsional, and curved physical properties of single-molecule DNA using optical tweezers and magnetic tweezers, and then completed the first experimental observation of ecological game theory in history with yeast systems during the MIT postdoctoral period
.
Jeff Gore began building the Systems Ecology team at MIT as an independent PI in 2010 and became full professor in 2021, where he is working to study how interactions between cell monomers emerge from the complex ecological and evolutionary dynamics of
microbial communities 。 Gore Lab combines experiments and theories to make representative achievements in ecological dynamic bifurcation, multiple steady states, symbiotic relationships, multi-species assembly, biodiversity and stability, and phase transitions of complex ecological networks, and his papers have been published in the journals
Science, Nature, Cell, PNAS, Nature Ecology & Evolution.

Hu received his B.
A.
from Tsinghua University's Qian Xuesen class, his Ph.
D.
under the supervision of Professor Jeff Gore of MIT's Department of Physics, and is currently an independent postdoctoral fellow at
MIT Physics of Living System.
His main research interests are the complex behaviors that emerge in multicellular living systems, in particular the discovery of coarse-grained parameters that effectively predict the spatiotemporal evolution of complex systems, thereby reducing the degree of
freedom required to quantitatively describe multicellular complex systems.
A number of research results led by Dr.
Hu Jingliang have been published
in Science, PNAS and other journals.

Thesis Links: