Roche’s Institute of Human Biology: The scientists transforming the future of drug discovery with human model systems
:quality(90)/)
With the inauguration of its new research building in Basel, Roche’s Institute of Human Biology (IHB) gains a new home—and drug discovery and development will gain something rarer: experimental systems built to behave like the human body.
In essence, Research and Development (R&D) is about translating science into medicines. The only issue is the translational gap — when promising discoveries stall between the laboratory bench and the patient’s bedside. A molecule behaves perfectly in a preclinical model, yet may fail in humans. Biology, it turns out, sometimes fails to translate neatly across species. Across the industry, organisations are trying to improve the success rate of projects.
The Institute of Human Biology (IHB) is poised to transform the future of R&D at Roche by focusing specifically on increasing that success rate, which is expected to help close that translational gap and bring life-changing medicines to future patients in need faster than ever before.
A key set of new tools in development at IHB is called human model systems: experimental platforms designed to reproduce key aspects of human biology outside the body. Some grow from adult stem cells into organoids—tiny living tissue structures that resemble organs such as the lung, liver or intestine. Others grow inside microengineered chips that recreate the mechanical forces and fluid flows of the body. Still others exist only in code, where computational and AI models simulate healthy tissues or disease.
The ambition is straightforward: bring the patient aspect into drug discovery from the very beginning. Better predictions earlier in research could mean fewer failures later—and more medicines reaching patients faster.
IHB is housed in a newly renovated facility in the Roche global headquarters in Basel, bringing together its three research cores: Exploratory Biology, Computational Biology and Translational Bioengineering — blending the curiosity of fundamental science with the practical demands of drug discovery.
The work begins in the Biology core, where scientists chase the "why" of disease with the tenacity of detectives.
Meet three scientists whose scientific contributions are shaping the future of research: Marie Bannier-Hélaouët, who explores the intricate workings of living systems; Christoph Harmel, who deciphers biology through computational patterns; and Saba Rezakhani, who architects human model systems with advancing bioengineering techniques. These experts embody the ambition and ingenuity that drive IHB’s mission forward.
The interpreter of living systems: Marie Bannier-Hélaouët
In one of Marie’s experiments, the cells begin to swell. They are part of a tear gland organoid. When stimulated, it produces tears, and the cells stretch. “They’re doing exactly what they’re supposed to do,” she says. “But without an outlet, the fluid remains trapped.”
Marie’s work begins with a deceptively simple question: how do organs actually function?
She focuses on epithelial tissues— thin layers of cells that line organs and carry out much of their essential work. Marie studies these systems in close collaboration with the Ophthalmology team in Pharma Research and Early Development (pRED).
But function does not arise from cells alone. “It also depends on the signal that cells receive,” she explains. By combining epithelial organoids with nerve cells, she looks at how signals can change what a tissue does.
To build these systems, Marie, a biologist, works with bioengineers, who help assemble different cell types into controlled environments. Computational biologists analyse the resulting data, comparing the behaviour of these models to real human tissue.
But before a system can be engineered or simulated, it must first be understood—how cells organise, how signals shape behaviour, and how disease emerges.
:quality(90)/)
For Marie, that connection between discovery and application is what makes the work meaningful. “In academia, you try to understand,” she says. “Here, when you understand, you know it might improve lives.”
But understanding comes with complexity. The interactions she studies produce vast amounts of data. Making sense of it is the role of the Computational Biology core—the bridge between the petri dish and the silicon chip.
The codebreaker of biology: Christoph Harmel
His colleagues study human model systems like organoids under microscopes, Christoph explores them through data patterns. A computational biologist who has been with the Institute of Human Biology from the beginning, his goal is simple to describe and difficult to achieve: translate living biology into something that can be measured, modelled and predicted.
:quality(90)/)
“In a way, I try to turn organoids into numbers,” he says. On his screen, they appear as code and data—representing cell arrangements, shapes and signals that reveal how the system behaves. His work begins by mapping the identity and arrangement of individual cells within a tissue. He then traces how molecular signals, metabolism and structure interact—until complex phenotypes—observable traits—begin to emerge.
From there, models emerge. While experiments can take weeks, simulations can test a hypothesis in hours. Increasingly, artificial intelligence helps detect relationships in the data that human observers might never notice. Experiments, by nature, can only sample a fraction of what's possible; simulations help fill the gaps—pointing to where in the experimental space it's most valuable to look next.
For Christoph, discovery follows a simple rhythm—define a problem, solve it, then move to the next. Much of the satisfaction, he says, lies in that flow.
“In a way,” he adds, “ideas grow here at IHB much like the systems themselves.” Both begin as small concepts, then evolve, combine with others and gradually become part of a larger framework that may one day help patients.
The final piece of the puzzle is the Translational Bioengineering core. If Exploratory Biology asks why disease happens, and Computational Biology helps explain how, Bioengineering builds the experimental systems that turn those insights into tools for drug discovery. One of the scientists doing exactly that is Saba Rezakhani.
The biological architect:Saba Rezakhani
In her sunlit laboratory on the sixth floor of the new research labs, Saba observes how human stem cells slowly organise themselves into — miniature kidney-like structures living systems that recapitulate key features of the real organ. “It’s like watching biology build itself,” she says.
As a child, Saba imagined becoming a detective. Today, she applies that same instinct for solving puzzles to biology. Originally trained as a chemical engineer, she was later drawn toward bioengineering—approaching complex biological problems with precision.
At IHB, she can pursue questions knowing the answers might shape real drug discovery. Using human stem cells, she grows organoids roughly half a millimetre across, about the size of a salt grain.
“A kidney isn’t just one cell type,” she explains. “Unlike simpler models that only focus on a single component, our organoids integrate the diverse cell populations needed to truly mimic the organ. If we want to study a disease like diabetic kidney disease, we need to ensure the system includes the specific cells responsible for glucose uptake alongside their natural neighbours.”
:quality(90)/)
Her long-term vision is ambitious. “One day, human model systems could replace many animal experiments,” Saba says. “They’re closer to human biology—and they can help us make better decisions much earlier.”
If that ambitious vision is realised, drug discovery could become faster, more precise—and far less dependent on trial and error.
Today, however, animal studies remain essential to ensure safety. Human model systems still need to prove themselves. Closing that gap between vision and reality is one of the central challenges in modern drug discovery.
What is changing is how discovery happens: To watch biology as it builds itself. To connect the systems where function emerges. To read it as it takes shape—and to understand it early enough to change what happens next.