The Kavli Network Effect: Christoph Kirst

Over the past quarter century, The Kavli Foundation has partnered with scientists at the forefront of discovery. Through 20 international Kavli Institutes — along with fellowships, funding programs and other collaborative initiatives — we see brilliant minds connect with one another, exchange ideas and explore the deepest questions about the universe, from the vastness of space to the workings of the human brain.

This series, The Kavli Network Effect, explores stories of scientific discovery, connection and collaboration by highlighting scientists who are part of the Kavli ecosystem. Each profile traces how encounters throughout this community have shaped individuals’ careers, leaving lasting impacts on their approaches to research, leadership and mentoring.

Part 1 profiles a neuroscientist’s ambitious investigation of the brain as a whole computational phenomenon.

For Christoph Kirst, his first encounter with the Kavli world came through one of the celebrated workshops hosted at the Kavli Institute for Theoretical Physics (KITP) at the University of California, Santa Barbara. These gatherings, lasting weeks to even months, bring together a select group of international researchers to tackle a compelling topic from a multidisciplinary and collaboration-sparking perspective.

Those early KITP workshops foreshadowed the kind of science that Kirst would pursue that thrives at the intersection of disciplines. Moving from theoretical physics to neural systems and computational neuroscience, he has made a practice of bridging fields. Today, as an assistant professor at the Kavli Institute for Fundamental Neuroscience (KIFN) at the University of California, San Francisco (UCSF), about 300 miles north of Santa Barbara, his research harnesses insights from mathematics, physics, computer science and neurobiology to understand how the brain computes and generates thought. An apt term for this confluence is “integrative neuroscience.”

Kirst began his studies in Germany, looking to merge interests in mathematics and physics into a profound investigation into the most complex object known — the brain. After a colleague told him about a KITP workshop in biophysics, Kirst reached out to Boris Shraiman, a biophysicist at KITP, who invited him to attend — and he loved the experience.

“You get a huge number of experts in that field, all leading people, and you can discuss with them one on one, and they give talks, and you can interrupt any time with questions,” Kirst describes. “It’s a great format, a really amazing place to learn about certain subjects, and also to meet people.”

One of the experts, Eric Siggia, whom Kirst serendipitously met at the workshop encouraged Kirst to apply for a three-year independent postdoctoral fellowship at Center for Physics and Biology at The Rockefeller University. While there, Kirst spearheaded development of ClearMap, a software program for analyzing images of brain tissue that has been treated to be transparent and thus able to be physically mapped. The Kavli Foundation later supported Kirst and his colleagues’ efforts through the Neuro Open Source Software program.

As this fellowship position wound down, Kirst then successfully pursued a Kavli Physics Fellowship at the Kavli Neural Systems Institute, also at Rockefeller. Doing so allowed him to mature ClearMap and other theory-driven projects investigating flexible information processing in neuronal networks. “The Kavli role put me in a much better position to apply for professorships,” Kirst says.

All the while, Kirst kept attending KITP workshops. At one about auditory processing, Kirst met a professor at UCSF who informed him of a job opening in computational neuroscience at UCSF’s Kavli Institute. Better still, the role carried a joint appointment with the Lawrence Berkeley National Laboratory, where Kirst could team up with engineers and data scientists to bring quantitative tools to bear on the conundrum of the brain.

Given his diverse academic background and approaches to neuroscience, Kirst has cherished the range of disciplines and topics he has encountered across the broader Kavli network, which have offered him unique cross-disciplinary opportunities that have shaped his approach to brain research.

In a full-circle moment, Kirst is now in discussion with KITP about developing a course on neuromorphic computing, an approach that draws on the human brain’s organization and function to design more efficient hardware and software. The brain, roughly 3 pounds of “wetware,” performs complex computations using a mere 20 watts of power—less than most lightbulbs. As the massive energy needs of data centers for AI and other technologies rise, breakthroughs in computational efficiency are needed. Neuromorphic computing may point the way forward.

At KIFN, where Kirst now also serves as a steering committee member, he continues to keep an open mind, receptive to insights from other realms of science. “If you work on something and then you get stuck, suddenly you can be inspired by tools and techniques that you’re working with in these other fields,” says Kirst. “You might never have thought about that way forward if you were just stuck within that single problem by itself.”