Many proteins with biotechnological and pharmaceutical applications are difficult to produce at scale or lack sufficient stability — driving up production costs and ecological footprint. Our new project tackles this challenge head-on using machine learning-based computational tools to predict targeted amino acid substitutions that boost protein stability and production yield, without compromising function or structure.

🔬 Using two model proteins as testbeds, we will benchmark different computational design strategies and validate the most promising ones experimentally in bacteria — testing for stability, activity, and beyond. Building on these insights, our goal is to establish a general, broadly applicable workflow for protein sequence optimization. Stay tuned for updates as the project unfolds! 🚀

It explains and highlights features and problems of tools for Structure Prediction, Enzyme Discovery and Classification as well as Enzyme engineering.