Does evolution have a direction? The current consensus answer among biologists is, perhaps unintuitively, no. However, all species have been evolving for the same amount of time, which makes it hard to detect a progressive direction of evolution, even if one existed. We exploited the fact that protein-coding genes continue to be born de novo from junk DNA, which allowed us to compare recently born gene families to ancient proteins that genuinely are “more evolved”. In preliminary work, we found what appears to be a striking directionality in the evolution of the structural properties of proteins. We found that young proteins use a primitive strategy to avoid protein misfolding, while old proteins use a much more subtle strategy, suggesting a progressive shift in strategy over billions of years.

While these preliminary data are highly suggestive, we have not yet ruled out alternative explanations. Specifically, proteins born at different times might have had different properties from the outset because the conditions of life have changed dramatically over time. A second alternative explanation is that it may be the preferential retention of gene families with certain properties, even in the absence of the hypothesized directionally-biased process of descent with modification, that drives our preliminary findings. We will conduct a variety of sophisticated analyses of existing datasets to dissect the forces behind our findings, which we strongly expect to validate the first-ever arrow of evolutionary time over these alternative explanations. Mathematical theory will be used to explore the general conditions under which long-term directional evolution is expected to arise. The primary deliverables of this proposal will be the publication of peer-reviewed scientific and philosophical papers, and the organization of a conference on the science and philosophy of directionality in evolution.