​On Sunday, 7 September 2025, I took part in the Open Day (Tag der Neugier) at Forschungszentrum Jülich, which welcomed more than 22,000 visitors to its 1.7 km² campus. From 10:00 to 17:00, the event offered a wide range of opportunities to discover current research in areas such as hydrogen energy, climate science, quantum computing, and the bioeconomy.
 
As part of the programme, I contributed to several hands-on, interactive activities. These included a computer game illustrating the complexity of protein folding, creative activities for children like coloring biomolecule and cell-structure printouts, and building simple 3D virus models, an AI-based game that generated imaginative images combining proteins and animals, and a tool that translated visitors’ names into protein sequences. These activities provided an accessible and enjoyable way for participants of all ages to engage with science.
 
Beyond my own involvement, the event featured a wide range of scientific demonstrations and exhibits across the campus. Visitors had the chance to explore research on hydrogen as a clean energy carrier, advances in quantum technologies, and new approaches to climate and environmental challenges. Interactive displays explained how brain research and neuroscience are helping us understand health and disease, while labs presented insights into the bioeconomy and sustainable materials. Many stations encouraged hands-on participation, allowing guests to experience firsthand how cutting-edge methods—from supercomputing to imaging techniques—are driving innovation at Forschungszentrum Jülich.
 
Looking back, it was rewarding to see how scientific concepts could spark curiosity in both children and adults. Contributing to this exchange reminded me of the value of making research approachable and inspiring for a broader audience.

In my previous blog post, I discussed some scientific insights I obtained during my time at the European RosettaCon, and in my first blog post, I broached the topic of industrial research compared to academic research. In this third and final blog post, I would like to bring both of these experiences together and share some insights that were presented at the European RosettaCon by a panel of academic and industry researchers regarding the differences of working in academia, “big pharma”, and startups.

One of the experts on the panel revealed how they felt that the main advantage of working in a big pharmaceutical company is that you can focus on a narrow field of your expertise. That is to say, they can be the head and main responsible for, for instance, computational design and work together with the head and main responsible for medicinal chemistry. This contrasts with their experience outside of the pharmaceutical industry, where they had to take leading roles in both computational design and medicinal chemistry. In their view, the opportunity to be part of a team working towards the same goal with multiple department heads is incredibly powerful.

The panel then went on to discuss how they experienced that small biotech companies can have only one method or target they work on. Consequently, if new information comes out about that target, the company may need to rapidly adapt to stay competitive. In big pharmaceutical companies, however, especially from a computational perspective, one typically works on several targets at the same time, and if new information comes out about one specific project, they can just decide to pause that project for weeks or years and put more focus on another project to stay competitive.
​
Finally, the panel discussed the advantage of academic research in the ability to work on a specific topic of interest that does not need to have an immediate business impact. However, they then shared that it is also possible to do such work in industry. Indeed, companies can similarly apply for external grants to get a PhD or a postdoctoral researcher to work on a topic that does not necessarily have to align with the immediate corporate goals. In fact, this is exactly what happened in my situation, as J&J became a beneficiary in the ALLODD consortium to host me as a PhD student to work on simulation-based methods for cryptic pocket discovery. Of course, there are many other aspects to discuss when it comes to comparing academic and industry research, but I found the insights shared by this panel quite instructive and wanted to share them in this final post. One thing is for sure: whether in industry or academia, there is always interesting science to be done and exciting discoveries to be made. Thanks for taking the time to read my blog posts and take care!

In the Autumn of 2024, I had the pleasure of going on a secondment to Novo Nordisk in Copenhagen, Denmark. With most of my PhD work focusing on molecular dynamics simulations to search for (cryptic) small molecule binding pockets, it was truly an insightful shift of gears to get some exposure to larger modalities and generative artificial intelligence (AI) methods. The idea is straightforward: given a target protein, generate the backbone of a protein binding partner, then generate a multitude of sequences that should be able to adopt this backbone fold, and finally select designs based on your favorite scoring function. In practice, the process is naturally more involved, but it was still remarkable to see just how powerful this approach can be. This not only became clear during my stay at the company, but to finish off my secondment, I also got to attend the European RosettaCon.

The work presented by academic and industry groups alike at this conference was a true showcase of the value that generative AI can bring. First, several works showed that de novo protein design can generate enzymes that not only match the catalytic efficiency of their natural counterpart but even exceed it by orders of magnitude. Then, there were examples of methods that can design functional protein binders in one shot and pipelines to rapidly design antibodies. A final highlight for me was a deep dive into how AlphaFold2 generates structures. This work indicated that learning how to generate protein structures with reasonably accurate geometries is very fast, requiring only about 10% of the total data available for model training. However, it takes up to about 90% of training data before the model can assign an accurate confidence score that correlates with structure quality. In other words, the vast majority of training data is not required to achieve high-quality results but rather to get an accurate understanding of the quality of the results.
​
Altogether, my experience in Denmark has been instrumental in widening my perspective on the field of therapeutics discovery from simulation-based approaches and small molecules to generative AI-based protocols and protein design. It is interesting and encouraging to see the impact that computational work can have across the board in the pharmaceutical field. Whether the aim is to develop a new binder for a hidden pocket that only appears in rare protein conformations or to obtain protein-binding proteins, there are always actionable insights to be gained from computational investigation.