July 31, 2019

Predicting Neighbor-Dependent Microbial Interactions

A new microbial network inference method reliably predicts interactions dependent on neighboring organisms

Model of MIIA Framework

A new microbial network inference method reliably predicts interactions dependent on neighboring organisms.

The Science

Soil microbial communities are made of networks of interacting species that dynamically reorganize in a changing environment. Understanding how such microbiomes are organized in nature is important for designing or controlling them in the future. Meanwhile, soil ecologists know well that some microbial interactions change because of neighboring species—so-called context-dependent interactions. Yet no theoretical framework, until now, has been available to address this phenomenon. In a new paper, researchers from PNNL’s computational biology and bioinformatics group proposed a method that reliably predicts how the modulation of interactions occurs in microbial communities subject to membership changes. The method, called minimal interspecies interaction adjustment (MIIA), addresses the problem of such context-dependent microbial interactions by accurately predicting dramatic shifts in microbial control of carbon and nitrogen cycles in the environment.

The Impact

By providing unprecedented predictions of neighbor-dependent interactions, the new computational method significantly improves our understanding of microbial network reorganization. This method will also enable the rational design and engineering of microbial consortia and natural communities.


Microbial community dynamics in soil and other habitats involve nonlinear interspecies interactions, so these dynamics are notoriously difficult to predict. Yet understanding how such microbiomes are organized in nature is necessary for designing them (such as for biofuel production) and for controlling them—for example, as a way to assure that soils do not emit too much carbon into the Earth’s atmosphere. Meanwhile, ecologists know that interactions in microbial communities are influenced by neighboring species. Until now, however, there has been no theoretical framework for predicting such context-dependent microbial interactions.

The research was motivated by the following fundamental ecological questions: How are interspecies interactions modulated by shifts in community composition and species populations? And to what extent can interspecies relationships observed in simple cultures be translated into complex communities?

The researchers addressed these questions by demonstrating that MIIA enables microbial interactions in binary cultures to be translatable into complex communities. The researchers also demonstrated the utility of this method in designing and engineering microbial consortia. In this regard, they found that microbial interactions can be significantly modulated when perturbed by a small number of neighboring species—but that the level of modulation diminishes as the number of new neighboring species increases.

This work, the authors say, can also be applied to questions of community ecology beyond microbes. It may provide a theoretical platform for better understanding all biological interaction systems, including human interactions.

PNNL Principal Investigator

Hyun-Seob Song
Pacific Northwest National Laboratory


This research was supported by the U.S. Department of Energy (DOE) Office of Biological and Environmental Research (BER), as part of Foundational Scientific Focus Area (SFA), Soil Microbiome SFA, and Subsurface Biogeochemistry Research (SBR) SFA at the Pacific Northwest National Laboratory (PNNL).   

Revised: July 31, 2019 | Published: August 1, 2019

H-S. Song, J-Y. Lee, S. Haruta, W.C. Nelson, D-Y. Lee, S.R. Lindemann, J.K. Fredrickson, and H.C. Bernstein,”Minimal Interspecies Interaction Adjustment (MIIA): Inference of Neighbor-Dependent Interactions in Microbial Communities,” Frontiers in Microbiology (2019). [DOI: 10.3389/fmicb.2019.01264]