This past fall, faculty and graduate students from the Department of Statistics, and other units at UW–Madison, participated in an intensive semester-long research program hosted by the Institute for Computational and Experimental Research in Mathematics (ICERM) at Brown University, entitled Theory, Methods, and Applications of Quantitative Phylogenomics. Among the program’s organizers were two UW–Madison faculty members, Professor Cécile Ané of the Departments of Statistics and Botany and Professor Sebastien Roch of the Department of Mathematics.
Between September and December the ICERM program brought together more than 130 leading statisticians, biologists, computer scientists, and mathematicians from around the world to address a critical question facing their fields: How can researchers make sense of the unprecedented amount of genetic data now available, spanning more species than ever before?
“One plus one equals more than two, especially in a program like this one that encourages a collaborative mindset,” Ané said. “We built on each other’s ideas to make progress that we would not have made separately.”
Since its inception in 1960, the Department of Statistics has prioritized interdisciplinary projects, bringing statistical expertise to other units across campus and beyond, through initiatives like the Statistical Consulting Group. The involvement of Statistics faculty and students in the ICERM program is the latest example of the department’s highly collaborative approach to research and problem-solving.
Big data questions
Ané stressed that impactful breakthroughs in quantitative phylogenomics—a field harnessing statistics, mathematics and computing to map out how different species are related—can happen when researchers from diverse backgrounds convene and share ideas, as they did at ICERM.
“We combine techniques from algebraic geometry, statistics, graph theory, algorithmic development, and knowledge about biological applications,” she explained. “Then, combine this with software development, and we go from theory to methods and software that evolutionary biologists can use on real data. That’s very exciting.”
Several other UW–Madison-affiliated researchers also participated in the program, including Benjamin Teo, a Statistics PhD student working with Ané; Sungsik (Kevin) Kong, a former Wisconsin Institute for Discovery (WID) researcher and current postdoctoral researcher at ICERM; Max Hill, a recent PhD graduate from the UW–Madison Department of Mathematics; and Joshua Justison, who will join WID in 2025 as a postdoctoral researcher working with Claudia Solis-Lemus, an assistant professor of plant pathology who earned her PhD from the Department of Statistics in 2015.
Hill explained, in the program’s newsletter, that statisticians are important contributors to quantitative research on evolutionary history because that history inevitably involves “randomness”, a concept statisticians study closely. Chance comes into play, for example, in genetic mutations, in which spontaneous errors alter the makeup of DNA. These random mutations have far-reaching impacts, determining physical traits, affecting the likelihood of developing diseases, and ultimately contributing to the creation of new species—Homo sapiens included.
Pushing frontiers
During the ICERM program, Ané and colleagues were able to restructure a key coding package, in the Julia programming language, that allows extremely efficient analysis of large amounts of genetic data. “I achieved the major goal, with others at ICERM, of refactoring the open-source Julia package PhyloNetworks into a smaller, lighter package,” she said. This big step forward, she said, will help researchers more effectively “solve various statistical analyses involving evolutionary trees and networks.”
“We now have a modular ecosystem that is easier to maintain—and grow,” she said, referring to the ability for any researcher to contribute new code to the Julia package. Anyone can now “use features already available and develop new features they need for their own research,” Ané noted. “Many current contributors [to PhyloNetworks] were present at ICERM, as well as researchers wanting to contribute new features. Being there together in person made this much easier to accomplish.”
Additionally, Ané said that the program enabled new team project ideas to emerge and grow in the coming months and years. “I started many exciting new collaborations, with researchers from the US and all over the world, [including] the Netherlands, the UK, Spain, and Canada,” she said. “We will continue to work remotely to finish these exciting projects.”