Tracy Heath received a grant from the National Science Foundation (NSF) to improve a software platform that enables research on the evolutionary relationships of species in the tree of life. These relationships are depicted as branching tree diagrams called phylogenies.
Heath, assistant professor of ecology, evolution and organismal biology, researches the development and application of computational methods for understanding evolutionary processes in a phylogenetic context.
An NSF grant for more than $400,000 will enable her and a team of developers to improve the stability, usability and speed of RevBayes, a statistical software platform that uses data to infer phylogenetic relationships.
RevBayes was originally designed and developed by Sebastian Höhna, professor of evolutionary biology at Ludwig Maximilian University of Munich; Fredrik Ronquist, professor of bioinformatics and genetics at the Swedish Museum of Natural History; and John P. Huelsenbeck, professor of integrative biology at the University of California, Berkeley; as a rewrite of MrBayes, a commonly used phylogenetic software using Bayesian statistical analysis. Through the process, RevBayes evolved into a more complex and more flexible software than MrBayes.
Heath got involved with the project in 2011 while a postdoctoral fellow in the lab of Huelsenbeck, with whom she shares the grant for the project.
“I’m really interested in technical things associated with software development and inference, which is statistical models that are motivated by biological processes,” Heath said. “As an interdisciplinary scientist the biological questions are really what motivate me.”
The software is designed to answer nearly any question an evolutionary biologist might ask, such as how different species are related, how mass-extinction events shape species diversity, or what the pattern of evolutionary rate variation across the genome is.
To answer those biological questions, researchers need software that can model their questions. Without the research tools, said Heath, “we wouldn’t have really good, robust understanding of the questions that we pursue in biology.”
While many tools exist for this work, few have the development support that RevBayes has with this grant. Heath’s work will improve the speed and efficiency of analysis, the user interface and the ability to work with other software programs.
The funding will help the tool become more robust for users with varied applications. RevBayes has its own interpreted programming language, which is what gives it the flexibility to write new models. The graphical user interface is also being developed in innovative ways not present in other phylogenetic software.
Some of the funding will also go to hosting workshops for training both users and developers, including students who are beginning research in computational phylogenetics.
“It has a lot of potential to provide a research analysis environment for biologists who ask questions in ways that they can’t from other software,” she said. “I think that the potential is going to make it really useful for the community.”
Other collaborators with RevBayes include Michael Landis, postdoctoral research fellow at Yale University; Bastien Boussau, National Center for Scientific Research researcher at the Laboratory of Biometry and Evolutionary Biology in Lyon, France; Nicolas Lartillot, research director at the Laboratory of Biometry and Evolutionary Biology in Lyon, France; Walker Pett, postdoctoral researcher in ecology, evolution and organismal biology at Iowa State University; and William Freyman, postdoctoral researcher at the University of Minnesota.