Machine-learning driven encapsulation of creatine using biomolecular condensates

Overview

This innovative solution leverages the natural phenomenon of liquid-liquid phase separation to stabilize creatine in biomolecular condensates formed by intrinsically disordered proteins (IDPs). By using machine learning, we aim to predict and optimize the encapsulation of creatine within these membraneless compartments, thus significantly enhancing its shelf-life. This technology offers a controlled platform for chemical stabilization, particularly benefiting industries reliant on creatine's stability, such as sports nutrition and pharmaceuticals.

Technical specifications

• Machine Learning Model: A predictive model designed to determine the degree of creatine enrichment in biomolecular condensates based on IDP sequences.
• Biomolecular Condensates: Formed through multivalent interactions between IDPs, these compartments are tailored to encapsulate creatine.
• Iterative Design Process: Involves creating a library of IDPs, screening them in silico, and refining based on empirical data to optimize creatine encapsulation.
• Long-term Stability: Focused on designing IDPs that provide a long shelf-life for creatine through precise sequence engineering.

Technology readiness level

Currently, this technology stands at TRL 3. It involves active research and development stages, with ongoing validation efforts. The process includes building predictive models, screening potential IDP candidates, and empirical testing of creatine encapsulation, aiming to refine and improve the technology for practical applications.