Using math to make biological science measure up

Bartek Rajwa, a Purdue associate professor of computational life sciences, has a job that epitomizes the interdisciplinary nature of Bindley Bioscience Center:
“I try to link and translate biological problems into frameworks which are compatible with some kind of mathematical modeling,” Rajwa said. “My job is to be in between – the glue – between the people who do advanced AI and machine learning and statistical analysis and people who sit in the lab and try to run the experiments.”
Rajwa has used spectroscopy to authenticate cheeses – to determine, for example, whether something labeled Gruyère is really Gruyère. The issue is much bigger, of course. “Even if you are not into artisanal food which costs a lot of money, you can still be a victim of food fraud.” he said, noting that Increased importing of food has raised concerns about safety. The goal of this research: to create scientific “fingerprints” for food.
Rajwa has also studied how to determine leukemia patients’ responses to chemotherapy. Like the authentication of cheeses, this is an issue of both biology and math: “How do you represent the condition of the immune system mathematically so you could feed it into some kind of mathematical model that would give you a prediction?”
Rajwa is different from some of the Bindley staff in that he’s not just trying to help researchers use current instruments but is helping them devise new methods of scientific analysis.
“The most important part of Bindley operations are labs where you run the measurements,” Rajwa said. “… This is sort of a hardware-based operation. And I would say that 95% of Bindley activity is to provide expertise on how to use this hardware to perform the actual measurements. In this sense, my story would not be typical for Bindley … I try to help people who are in other colleges on the campus to build new measurement systems, new approaches. … My role is to be an enabler.”