Over the past millennia, artificial selection or breeding has allowed us to tame nature and greatly improve our lives. Artificial selection need not be restricted to plants or animals, but can be similarly applied to microbial communities, which we now know to provide huge benefits to our health and our environment. However, because microbial communities contain many co-evolving species, each made up of different individuals, selection at the community level often fails.
We have been studying why this happens and how to avoid it [1]. Based on this understanding, we then developed a novel selection method that we call “disassembly selection”, where we construct small bacterial communities from a defined pool of culturable isolates and subject them to several rounds of selection, where at each round, we disassemble the communities and reassemble them into new combinations that resemble the winning communities. We have build computer simulations, which have shown that this approach can be more effective at finding the best community compared to the state-of-the-art approach to community-level selection [2], and we have tested our approach experimentally in the lab [3]. Our laboratory experiments were applied to finding a community capable of degrading industrial pollutants called Metal Working Fluids. Over the course of 18 weeks, we were able to find a community that degraded 69% more pollutant than the original community we typically work with in the lab. These experiments taught us a lot about how species come together to solve this task and how to improve our method. This is an important avenue for future work in the lab.
References
1. F. I. Arias-Sánchez, B. Vessman, S. Mitri (2019) PLoS Biology (Primer).
2. B. Vessman, P. Guridi-Fernández, F. I. Arias-Sánchez, S. Mitri (2023) bioRxiv.
3. F. I. Arias-Sánchez, B. Vessman, A. Haym, G. Alberti, S. Mitri (2023) bioRxiv.