Claudia Solis-Lemus PhD’15 and her lab bring statistical rigor to far-reaching biological questions.
By Thomas Jilk
When Claudia Solís-Lemus PhD’15 began her Statistics graduate program, she was leaning toward focusing on financial statistics. Then she attended a guest lecture from Professor and current Statistics Chair Bret Larget, showing how probability models could uncover new patterns in the evolution of life, using DNA as data.
“Lightning hit my brain,” Solís-Lemus said. “It was the coolest thing I had ever heard in my life.” From that moment on, her career was set on a course toward advancing the intersection of statistics and phylogenetics, where creative approaches in data science can help reveal a fuller picture of the Tree of Life — the extraordinarily complex web of histories and relationships that constitute the living world.
Today, as an associate professor in the Department of Plant Pathology, Solís‑Lemus leads an interdisciplinary lab at the Wisconsin Institute for Discovery (WID), where her team models evolutionary dynamics, from plant and animal species to microbial communities inside the human body. The statistical and computational tools they develop enable scientists to more efficiently address consequential questions in public health, environmental and agricultural sciences, and any field where understanding life’s hidden histories can drive new breakthroughs.
“The ultimate goal is to reconstruct the Tree of Life, from the origin of life to the diversity we see today,” Solís-Lemus explained.
Her path from Statistics PhD student to principal investigator at WID reflects the Statistics department’s long tradition of training data-savvy researchers who bridge fields and solve real problems.
Scaling up biological discovery
Much of the Solís‑Lemus Lab’s work centers on building statistical and computational tools that can keep pace with the sheer volume of modern biological data. For example, Nathan Kolbow, a PhD candidate in the Statistics department’s Biostatistics option, focuses on building computationally scalable statistical models that capture phenomena like hybridization, gene flow, and other evolutionary events that make the Tree of Life look more like a tangled web.
“A lot of existing methods break down once you get past a few dozen species,” Kolbow said. “Our goal is to make these models usable at the scale biologists actually need.” For example, a recent project (with Kolbow as lead author) develops a new package in the Julia programming language to enable phylogenetic inference — the statistical process of identifying evolutionary relationships — on a larger scale than was previously possible.
Advances like these could allow researchers to answer big questions in statistically and computationally rigorous ways. Such questions include: How did major plant lineages diversify over time? How do viruses mutate and spread across populations?
“Our goal is to make these models usable at the scale biologists actually need.”
Nathan Kolbow
As the lab pushes forward, they’re helping scientists move from simplified evolutionary trees to richer and more realistic models of life’s complexity.
Microbes and human health
In addition to building computational tools to recreate the Tree of Life, the lab is also exploring the intricacies of the microbiome and its evolution. Collaborating with colleagues across medicine and genomics, Solís‑Lemus and her students develop statistical tools to understand how microbial communities influence their hosts, including through the gut-brain axis.
“We know humans have microbes in our gut,” Solís‑Lemus said, “and we’re studying how these connections are affecting, in particular, brain functions.”
Statistics PhD student Jiayang Wang works on phylogenetic network modeling and sees the interdisciplinary nature of the work as essential. “Genomic data are incredibly messy and high‑dimensional,” Wang said. “You need statistical and biological knowledge, and computational tools, all at once. None of them works on its own.
The lab’s recent projects in this area include creating a statistical model and accompanying R package to infer microbial interactions and their connections to both environment and host; and building a new R package for accurately building consensus microbial networks, which integrate results from multiple algorithms to construct a single, more reliable network map. These tools help researchers investigate links between microbes and health-related outcomes, while also supporting environmental and agricultural studies that depend on understanding microbial dynamics.
These efforts connect evolutionary modeling and human health, showing how statistical tools can illuminate biological dynamics in a data-rich world.
A culture of collaboration
Across all their projects, collaboration is the lab’s defining feature. Students like Kolbow and Wang work closely with partners in fields like plant pathology, ecology, genomics, and medicine, translating statistical ideas into tools that domain scientists can use. “I try to be very collaborative,” Solís‑Lemus said, “I love working with people who think differently from me; that’s where the interesting ideas come from.”
She continued, “Interdisciplinary work is not optional in this field. It’s the only way to make progress.” The graduate students echoed that sentiment. According to Kolbow, “The questions biologists ask drive the models we build. Working with collaborators helps us understand what’s biologically realistic.” Wang added that in the realm of microbes, “talking with collaborators helps us figure out which microbial relationships actually make sense.”
“Interdisciplinary work is not optional in this field. It’s the only way to make progress.”
Claudia Solís‑Lemus
In their commitment to building shared tools and working toward breakthroughs that benefit scientists and society, the lab is living out the Wisconsin Idea, one statistical model, one coding package, one innovative application at a time. The Tree of Life is full of mysteries, and the Solís‑Lemus Lab is working across disciplines to solve them.
