Gregor Horstmeyer, Co-Founder & CEO, Ramona

In an interview with Invest:, Gregor Horstmeyer, co-founder and CEO of Ramona, discussed the shift from animal testing to humanized in vitro models and the growing role of artificial intelligence in drug development. “The work of biologists is not going away. I just think that AI will continue to be an accelerant for scientists, drug developers, and people working on new therapies,” Horstmeyer said.

What changes over the past year have had the greatest impact on Ramona Optics, and how have they shaped your leadership priorities?

Two developments have had the largest impact on how we think about the future of the company. The first is the government’s push to move away from testing drugs in animal models and toward testing them in humanized in vitro models, meaning cells and assemblies of cells in a petri dish.

It is fantastic that this transition is happening and that it has bipartisan support. There are ethical reasons to reduce animal testing, but drug testing on animals also does not work well. A drug may work in a mouse but not in a human. Compounds may also produce toxicity in humans that was not observed in a mouse, dog, or nonhuman primate because humans metabolize therapies differently.

The FDA, NIH, and regulatory agencies outside the United States understand that these limitations make drug development slow and expensive. There is a real push to replicate human biology in a dish, test compounds using humanized models, and limit the amount of animal testing required. This is a significant tailwind for Ramona Optics because we make technology to measure living biology. We want to position ourselves in the middle of that ecosystem.

The second development is AI. We use these tools daily across the company, particularly to accelerate software development. The available tools are well-versed in coding languages and allow people to communicate with them in English. Work still needs to be done after receiving the output, but I view AI as an accelerant, not a replacement. 

I believe AI will become increasingly impactful in drug development and life sciences. However, I do not believe AI will discover future drugs on its own or that everything can be done through computer simulations. We are still naïve about how biology works, so ground-truth experiments with real biology remain necessary.

The work of biologists is not going away. I just think that AI will continue to be an accelerant for scientists, drug developers, and people working on new therapies.

What challenges in drug discovery inspired you to build Ramona Optics, and where do you see the greatest opportunities for innovation?

For much of the past 50 years, the early stages of drug development relied on measuring static biology. Researchers would freeze cells at a particular point, fix them in time, take images, and analyze those images. That was partly a result of what measurement tools could do and partly due to the difficulty of analyzing image data.

Images are incredibly rich, but humans have difficulty quantifying what they contain. We are good at making qualitative assessments. For example, a person driving can interpret the environment and respond to changing conditions, but that person would struggle to state precisely how many other cars are on the road.

Drug development requires quantitative information. Researchers cannot make decisions based only on whether something looks slightly better or worse. They need to know how many cancer cells died after exposure to one drug compared with another. A human looking at images containing 510 objects and 498 objects would not reliably identify the difference, but those numbers can be meaningful to drug developers.

What drove me to the company was that imaging and computational capabilities were both changing. We can now develop technology that captures a series of images or videos showing how living biology unfolds. We also have computational tools capable of analyzing static images, image sequences, and movies.

Together, these advances allow us to build tools that help researchers understand how drugs affect living biological systems, rather than relying solely on a fixed point in time from a nonliving sample. Ultimately, these drugs are intended to help people combat disease and live longer. That fundamental idea of building tools to probe living biology attracted me to this work.

How important is speed in biomedical research, and how can your technology help accelerate the process?

Biomedical research, particularly drug development, is notoriously slow. It is slow for good reasons. Researchers need time to understand the long-term effects of compounds, and scientists are naturally skeptical. We want them to probe exactly what is happening and understand how things work as thoroughly as possible.

However, drug development is also slow for reasons it should not be, including failures at different stages. A compound may look promising early, develop problems in an animal model, and require researchers to return to the beginning. It may later succeed in an animal but fail during the first phase of a human clinical trial.

We are interested in helping researchers reach those failure modes as quickly as possible so the iteration cycle can restart. That requires combining the appropriate technological systems with humanized cellular models to identify problems before they become late-stage failures.

If researchers work on something for five years and then encounter a failure, they have lost five years. If the failure occurs after one month, restarting the search is more feasible in terms of both time and cost. Researchers can fail several times, learn from those failures, and continue the campaign.

When a failure is not identified for five years, there may be only be one opportunity. So much money and time have already been invested that the failure often marks the end of the campaign.

What makes the Research Triangle attractive for companies developing advanced technologies and conducting research?

The best word to describe the region’s advantage is talent. There are not many places in the United States with the talent density found in the Triangle, and that density is a huge reason our team now stands at 26 full-time employees.

It begins with the academic institutions. UNC-Chapel Hill, NC State, Duke University, and Wake Forest are all within driving distance. In our case, Duke is about five minutes away. 

That academic engine helps attract companies to the region. Once companies arrive, they draw in additional talent as people relocate here. That concentration of talent is what excites me most about building a company in the Triangle.

What are your top priorities for the next few years?

Our first priority is continuing to expand our commercial base and get our technology into more laboratories. Our technology does not proliferate through a top-down process. It happens by working with scientists on the ground.

Scientists are the users and adopters. They ask the right skeptical questions and conduct the daily work with humanized in vitro models and compounds. They also provide guidance to company leadership about whether a compound should move forward or undergo additional testing.

We are continuing to invest in our ability to reach the right people when they are making decisions about adopting new technologies. That is probably my highest priority.

The second priority is continuing to invest in the technology stack that enables us to generate data and accelerate the AI-related process. We have an opportunity to position the company as a catalyst for a future in which wet labs remain part of the loop while AI accelerates the work.

That requires generating accurate, quantifiable data at scale that can be used for decision-making. Our measurements involve living biological systems, which we believe are among the most impactful measurements available during early drug discovery.

What role has the SBIR program played in supporting Ramona Optics and scientific innovation?

The Small Business Innovation Research program is a distinctive U.S. program for supporting early-stage innovation. It allows experts to approach the government with a difficult scientific problem, explain their proposed solution, and request funding to test whether the idea works.

Ramona Optics’ two other co-founders, Roarke and Mark, used this mechanism after completing their Ph.D. programs at Caltech and after Roarke took a position at Duke University. They could demonstrate their expertise, present a plan for addressing a hard scientific problem, and share their progress with the government.

These ideas are often too early and risky for traditional investors. The first steps are not about commercial progress but about proving or disproving an idea. The SBIR program provides support during that critical period.

Ramona Optics would not be where it is today without that mechanism. It is a catalyst for innovation in hard, deeply technical sciences and encourages research with a commercial goal. Researchers can work on a difficult problem while explaining how they would commercialize the solution if it succeeds.

There are avenues beyond traditional investment mechanisms for people interested in starting companies or becoming entrepreneurs. The SBIR program is one of those avenues, and people with difficult scientific problems should explore it.