Growing things at a large scale can be a finicky affair. Whether it’s a field of sweet corn, or a vat of insulin-producing cells, the smallest differences in the conditions they both grow in can make the difference between a successful harvest or a meagre one. For corn, that environment is soil; for cells, it’s culture media.
Fine-tuning culture media formulations can be especially challenging for scientists and biomanufacturers, as the cell types they’re interested in normally function as part of a bigger living organism, making them reluctant to thrive on the benchtop or in factory lines. Too often, finding the perfect culture media ‘recipe’ for a particular cell type involves many time-consuming, costly and frustrating rounds of trial and error. Yet, this effort is often necessary to pave the way for advanced therapeutics, alternative foods and other biotechnology products that could revolutionise our lives.
Having personally experienced the persistent setbacks caused by suboptimal culture media, Zach Pang, a Group Leader at the A*STAR Bioprocessing Technology Institute (A*STAR BTI) and Senior Scientist at the A*STAR Bioinformatics Institute (A*STAR BII), has made it a lifelong mission to make culture media development faster, smarter and more accessible for scientists and biomanufacturers.
As co-founder and Chief Scientific Advisor of A*STAR spinoff company AuctuCel, Pang is part of a team of entrepreneurial researchers that are bringing artificial intelligence (AI), computational modelling and bioprocessing together in pursuit of that goal. AuctuCel’s flagship technology, Fluxperior, combines mathematical modelling with AI to predict what cells need and optimise culture media formulations based on those needs, reducing development time and costs while supporting higher-yield and more scalable biomanufacturing.
In this A*STAR Research feature, Pang reflects on the origins of his mission to improve culture media development; the potential for models like Fluxperior to accelerate bioprocessing R&D across diverse industries; and the benefits and challenges of a career spanning both research and entrepreneurship roles.
1. What sparked your interest in changing the culture-media development process?
During my PhD degree, I worked with primary cultures of human endothelial cells; specifically, those isolated from donated umbilical cords and human aortas. These cells are notoriously difficult to grow, and they’re very particular about what culture media they grow in. In my first year of research alone, it took months of trying out different brands of culture media to find one that ensured the best cell growth—which also turned out to be the most expensive.
The thing about many culture media products is that they might be off-the-shelf, but they’re not one-size-fits-all. Different cell types from different sources cultured in different contexts will require different formulations. This present a big problem for companies involved in the biomanufacturing and biotech industries, which is why they spend lots of money and time developing their own formulations, or engage third-party vendors to do it for them.
2. What was the journey like to Fluxperior’s development and AuctuCel’s founding?
The Fluxperior platform started as a project by our former A*STAR BTI colleague, Meiyappan Lakshmanan, which we then applied to—and very successfully validated—in an industry project with a partner company. After that collaboration, we published an article about the work; our partner also launched their product in the market, crediting A*STAR in its development.
Later, in 2023, I was nominated by A*STAR BTI to attend the Gap Champion programme run by A*StartCentral (A*SC) and Singapore Biodesign. The programme arms scientists and business development officers with the business acumen to scope an industry-relevant grant proposal. While I was there, an A*STAR Innovation and Enterprise colleague asked me if I was interested in spinning off our technology into a company, considering it had already attracted industry interest. I took up the challenge, joined A*SC’s Essentials Bootcamp with my co-founders, and here we are.
We picked the name ‘AuctuCel’ by combining the Latin word ‘auctus’—meaning growth, enhancement and prosperity—with ‘cell’. Our goal is to help companies successfully grow their cells with minimal hurdles, allowing them to enhance productivity, which leads to prosperity.
3. How does Fluxperior work? What makes it a novel approach?
Fluxperior is a mathematical model coupled with advanced AI, allowing its users to design the best formulations for the culture media they need. By simulating intracellular metabolism, the model provides a glimpse of what happens in cells in culture, especially the metabolic bottlenecks present. Users can then simulate increasing or decreasing amounts of individual culture media ingredients to see what unlocks those bottlenecks, boosting cell growth and productivity.
4. What are some real-world applications of Fluxperior?
Our tool can be applied to any field that relies on biological cells grown in culture media. In the past, we’ve helped Ajinomoto develop a culture media product for Chinese hamster ovary (CHO) cells, which are commonly used by the pharmaceutical industry for antibody production. We’ve also applied our tool to projects involving vaccines, cell therapy and cultivated meat.
5. What is it like balancing your multiple roles in research and entrepreneurship?
It’s great and exciting to work on many things, but it can also be tough and mentally draining. While I’m not involved in AuctuCel’s operations today, in its early stages I helped advise it and kickstart its operations. Recruitment, fundraising, logistics and manufacturing all take time and effort; most of my evenings and weekends are spent talking to my co-founders and discussing strategy. At the same time, I also juggle my main job at A*STAR and adjunct professor positions at the National University of Singapore and Nanyang Technological University, Singapore.
However, I have no regrets; I’m loving it. Being involved in both research and entrepreneurships gives me a wide lens to see how the whole landscape works, from basic science discovery, translational research and software development, all the way to industry application and business development. I get to see the challenges that lie in different stages, which really broadens my perspective on how to do science.
In my opinion, I’d like to advocate for more scientists to be cross-trained—I don’t mean to move scientists away from the bench in favour of business relations, but to allow them to spend half their time engaged in business and commercial activities. I can say it hits the brain differently; I’ve even had to learn how to do accounting.
6. How is AuctuCel keeping ahead of the curve in AI and biomanufacturing?
The field of AI is growing crazily fast, but we don’t compete in the AI space itself. Since Fluxperior is a mathematical model, it has explainability; it lets us perform simulations and know the exact biological pathways that influence cell growth. AuctuCel has also pivoted into manufacturing and selling off-the-shelf culture media. To date, its product catalogue comprises over 200 products. The company aspires to be the next big biotech suppliers in Singapore and the region.
7. What advice would you give younger scientists with an interest in AI-forward bioprocessing?
Stay grounded in your domain—whether biology or biotechnology—and expand your expertise in AI. We need more AI scientists who also speak the language of biotech and bioprocessing!