Public summary Deliverable 3.4

The RADICALZ project is dedicated to accelerating the discovery and development of new enzymes. These enzymes are key ingredients for creating more environmentally friendly and efficient consumer products. Traditionally, finding and improving enzymes is a time-consuming and resource-intensive process, often involving many cycles of trial and error in the lab. Deliverable 3.4 highlights a major step forward in making this process smarter and faster. The project team has successfully integrated advanced Artificial Intelligence (AI) tools into an existing sophisticated software platform called 3DM, which is used by scientists for protein engineering. The new AI-powered features within the 3DM platform assist scientists at crucial stages of enzyme development: 1. Smarter starting point selection: One of the first challenges is choosing the best enzyme to start working with. A new AI tool now helps predict the stability of potential enzyme candidates (e.g., how well they can withstand heat, a critical factor for many industrial uses). This allows researchers to make more informed decisions and select promising starting enzymes, saving considerable time and effort downstream. 2. Intelligent design of new enzyme variants: Once a starting enzyme is chosen, scientists modify it to improve its properties. The AI now helps predict the likely impact of these modifications. For instance, it can flag changes that might render the enzyme non-functional. This "intelligent filter" helps researchers focus on creating diverse yet functional enzyme variants, reducing the number of unsuccessful experiments and guiding them towards more promising candidates. 3. Accelerated learning and optimization: The enhanced 3DM platform now allows scientists to easily upload their experimental data (e.g., how well a new enzyme variant performed). This information is then used to further "train" the AI models. The AI learns from these results and can then suggest even better enzyme designs for future experiments. This creates a powerful iterative cycle of design, testing, and AI-driven learning, speeding up the optimization process. The integration of these AI tools into the 3DM platform brings tangible benefits: • Reduced experimental workload: By predicting which enzyme candidates or modifications are most promising, the AI helps scientists avoid unnecessary lab work. • More focused research: Scientists can design "smarter" experiments, increasing the likelihood of discovering improved enzymes. • Faster development cycles: The entire process of finding, designing, and optimizing enzymes is significantly accelerated. In essence, the RADICALZ project is successfully leveraging the power of AI to provide scientists with cutting-edge, user-friendly tools. This advancement is a significant step towards the project's ultimate goal: the rapid development of novel enzymes for greener and more sustainable consumer products, benefiting both industry and society.