This week’s BEACON Researchers at Work blog post is by MSU graduate student Luis Zaman. Enjoy!
We’ve all been stuck in stand-still traffic on the highway. Slowly people start exiting to use a newly found alternate route. Unfortunately, this detour starts backing up as everyone else on the road decides to merge over four lanes of traffic. Soon, the alternate route is as backed up as the highway. In this situation, if everyone has the same distance to travel, the person who finds the least-used detour would arrive quickest. We call this situation “negative frequency-dependence,” where a frequently-used detour is worse than a rare one.
Negative frequency-dependence is also important in evolutionary biology, where rare types of individuals may be more successful than common types. The flu virus is a prevalent example: we build up immunity to the types of viruses we have encountered in the past, yet every year we are at risk of getting a new strain of the flu. The rare types of flu become successful (that is, they successfully infect many people) in a particular year because the general population does not have a built-up immunity to them.
I study what is known as “host-parasite coevolution,” or the reciprocal adaptation of hosts to their parasites and the co-adaptation of parasites to their hosts. In this form of antagonistic coevolution, negative frequency-dependent selection often maintains diversity. Think back to the congested highway: if the least traveled detour is best, then many different detours will probably be used at any one time – that is, there will be diversity of detours. Using a computer program called Avida in which digital organisms can evolve in different digital environments, we are able to study the effects of host-parasite coevolution in great detail, and with a level of control impossible in typical biological systems. In our newest paper (co-authored with Suhas Devangam, an undergraduate researcher working in the Devolab, Dave Bryson, and Dr. Charles Ofria), we explore how host-parasite coevolution maintains diversity in these digital organisms. Specifically we look at how host diversity is affected by the presence of parasites and the presence of mutations. Because Avida allows us to completely turn off mutations (which are an important part of biological evolution), we can take experiments where hosts and parasites are coevolving, and flip a switch to disallow any new variation. The following graph shows how, whether we leave the mutation switch on (a), or switch it off (b), host diversity is much greater in the presence of parasites.
A host that can escape infection will proliferate, but as it becomes more common, parasites will have increasing pressure to adapt to it. Hosts escape parasites by mutating into new species or types. Diversity is maintained when these types fluctuate back and forth as parasites target the most common hosts. That is, negative frequency-dependent selection imposed by the parasites maintains host diversity. This is true even after we stop mutations just as we would expect in our highway analogy: even if we were able to somehow stop any new detours from being found, drivers would still use a diverse set of alternate routes.
Now we know that coevolution with parasites increases diversity, and even maintains it in the absence of new mutations. But what are the effects of these diverse communities on the whole coevolutionary process? To begin answering this question, we need to “retrain our brains” to think in network contexts (as expressed by Ben Kerr at University of Washington). There are diverse communities of hosts and parasites interacting in complex ways, and coevolution is not strictly operating on any one interaction. Rather, coevolution affects the entire network. To get a glimpse into these complex networks, my collaborator, Miguel Fortuna (currently a post-doc in Simon Levin’s lab), turned data from digital host-parasite interactions over coevolutionary time into a movie.
In this movie, unique host and parasite species or types are represented as spheres. Green spheres are host types, and red spheres are parasite types. The size of a sphere represents the abundance of that particular type in the community. A link between spheres represents a parasitic interaction where the width of the green links depicts how popular a parasite is to the host, while the width of the red links depicts how popular a host is to the parasite. You can see that we begin with one type of parasite, and one type of host, but over time, both hosts and parasites diversify, and the interactions become very complex.
How coevolution shapes these networks, and how these networks shape the paths that coevolution takes are fascinating questions. The majority of my current work focuses on understanding host-parasite coevolution in communities with many interactions, and I hope my next post will shed some light on how coevolution in this context proceeds. For now, I wish you the best of luck at finding the detour least traveled.
For more information, contact Luis at luis.zaman at gmail dot com.