The Breakdown

• Research challenging a cornerstone of ecology could improve species conservation and help fight diseases.
• When species in an ecosystem cannot mix, the advantage that normally helps rare species survive may no longer work.
• Researchers hope this work can help increase biodiversity to aid with conservation and decrease cell diversity when treating diseases such as cancer.

 ***

A biological process long thought to protect biodiversity and help species coexist may actually threaten diversity when species are separated by natural landscapes, infrastructure or other barriers, according to new research from Carnegie Mellon University's School of Computer Science. The finding could help scientists better protect biodiversity — and may even offer new insights into how diseases like cancer evolve.

Researchers from the Ray and Stephanie Lane Computational Biology Department found that a long-standing idea about how ecosystems work may not hold up when habitats are fragmented. The concept, known as negative frequency-dependent (NFD) selection, suggests that rarer species tend to perform better than more common ones, preventing any single species from dominating an ecosystem. This dynamic gives rarer species a better chance at survival and has been used to explain patterns across biology, from bird species in Japan to cancer cells in the human body.

However, most ecological and evolutionary studies of NFD selection assume that populations mix freely rather than being divided by mountains, roads or other barriers. The new research shows that when species are separated, the advantage of being rare may disappear.

"My lab is trying to understand how the social and spatial structures in the world affect the evolution of pathogenic systems, from the microbiome to bacteria and viruses," said Oana Carja, an assistant professor in computational biology. "This work is important because we want to increase diversity in some cases, like animal species. But clonal diversity in cancer is harmful, and we want to minimize it. If we understand how spatial arrangement influences biodiversity, we can use that knowledge to make a real impact."

Carja said this research opens the door to more work into how space impacts biodiversity. Conservation groups spend millions of dollars building corridors to reconnect habitats, and these projects are even more important because of how separation impacts biodiversity. In the reverse, understanding how space impacts disease transmission could mean new avenues to treat illnesses.

NFD selection can help explain why one species doesn't take over if it is fitter. In smaller or well-mixed ecosystems, rarer species can persist because their rarity provides an advantage. Less common bird species might specialize in feeding on a specific insect in their local environment, for example. But in fragmented ecosystems, that advantage may break down.

The same concept applies to the human body. Inside a tumor, NFD selection allows rare cancer cells to survive and grow. In a small region of the tumor, rare cells and more common ones coexist, without the more fit cell taking over. But tumors often contain separated areas, like barriers of dead tissue. If these barriers didn't exist and the cancer cells could mix freely across the whole tumor, the local advantage of rare cells would diminish, making it less likely for them to persist throughout the tumor and decreasing the tumor's evolutionary potential.

"Earlier in my research, I saw a gap in how we understood spatial structure and these classic NFD selection results," Carja said. "In real systems we see in our world, from birds on islands off the coast of Japan to cancer cells inside the body, spatial structure is everywhere. These systems are not well mixed, and it was mind-boggling to me that this was not studied. Does spatial structure change these prior results examining negative frequency dependence?"

Carja and Ph.D. student Anush Devadhasan used mathematical models to see how rare species survive in fragmented ecosystems. They compared a species with two variants, one rarer and the other more common.

In one scenario, the rare species has an advantage or NFD selection is in place. In the other, all species were treated equally. In this initial experiment, the populations were separated and couldn't easily mix. Researchers found that, despite previously held assumptions, ecosystems with species treated equally maintained coexistence for longer than those with NFD selection. The local advantage NFD selection provides doesn't translate when species cannot easily mix. Even with more species added to the model, this result held.

Devadhasan then found a real-life dataset that mimicked the mathematical models and simulations: bird species in Japan's Ryukyu Islands. Researchers used this dataset to check their results. They found that NFD selection does operate locally at the specific island level, but the advantage rarer species have doesn't necessarily translate across the island cluster.

Moving forward, the researchers hope to develop tools that help build corridors between fragmented ecosystems, promoting biodiversity and species coexistence.

The paper was published in the Proceedings of the National Academy of Sciences.