Computational Challenges from the Tree of Life

A phylogeny is a reconstruction of the evolutionary history of a group of organisms. Phylogenies are used throughout the life sciences, as they offer a structure around which to organize the knowledge and data accumulated by researchers. Computational phylogenetics has been a rich area for algorithm design over the last 15 years. The biology community has embarked on an enormously ambitious project, the assembly of the Tree of Life—the phylogeny of all organisms on this planet. This project presents a true computational grand challenge: current phylogenetic methods can barely handle a few hundred organisms, yet the Tree of Life has an estimated 10–100 million organisms. Thus, while data collection and cataloguing is one of the major tasks, algorithm design and engineering is a crucial enabling component of the project. In this paper, I briefly introduce the principles of phylogenetic analysis, then discuss the computational challenges posed by the Tree of Life project, from the design and validation of computationally useful models of evolution to the actual computation and assessment of the Tree of Life itself.