In my job as a researcher I need to keep up with the latest developments in my field, but I also like to keep an eye on broader topics. The other day I came across a neat preprint on arXiv.org, in which the authors have shown a possible explanation for the curious lack of diversity amongst influenza strains around the world. This paper has several of the elements that I think sum up what Complex Systems research is about, and why it is so exciting:
A tricky problem, but one that relates to our personal experience. The judicious use of mathematical modelling and simulation to get a handle on the key elements of the puzzle, and come up with insights that otherwise would not have been possible.
The reason we keep getting the ‘flu is that the virus evolves rapidly, so the immunity we have from getting last year’s ‘flu isn’t that effective against this year’s. But if the virus is changing so rapidly, why don’t we see a huge diversity of different ‘flu strains, branching out in different directions? What we see instead is a “serial replacement of strains”, with successive strains cutting a nearly linear path through the space of antigenic possibilities. By carrying out computer simulations of influenza evolution, the authors have come up with an answer: our own immunity to previous strains sets up a chasing selection pressure. The most successful mutations would be the ones moving the furthest away from previous incarnations, continuing in exactly the same direction as previous changes. Mutations in any other direction are less likely to succeed, and those that lead back in the direction of earlier strains are doomed, as we already have a good immunity to them.
So far, so great, but the paper also dangles a carrot for a much bigger possibility. The evolution of the strains in their simulations was highly predictable for about the first year. If this carries over to the real world (and this is still a really long way off) it might mean we could make an educated guess at what strains of influenza will be around next winter. Imagine being able to get a ‘flu shot, not for this year’s ‘flu, but next year’s too. That would really leave the virus with nowhere to run.
Canalization of the evolutionary trajectory of the human influenza virus by Trevor Bedford, Andrew Rambaut and Mercedes Pascual