mRNA therapeutics have revolutionized the development of biologic products, yet their promise as a universal platform is contingent on shifting from empirical trial-and-error to predictive, computation-driven design.

Current pipelines are often hindered by lengthy screenings based on unreliable experimental models. To address this, diverse AI strategies—including Transformers, Graph Neural Networks (GNNs), and Generative Deep Learning—are being deployed to navigate the vast combinatorial space of mRNA and delivery systems.

In this special session, we cover three critical frontiers, and their integration into multi-objective optimisation strategies:

• Construct Optimization: Multi-objective design of UTRs, coding sequences, and poly(A)-tails to balance mRNA stability with cell-specific translation efficiency.
• Design for Manufacturability: Computational strategies to identify and mitigate sequence-dependent bottlenecks in both batch and next-generation continuous-flow manufacturing.
• Formulation Engineering: Data-driven and physics-based methods to optimize Lipid Nanoparticles (LNPs), targeting improved biodistribution, specific tissue delivery, and optimal pKa profiles.

These approaches aim to establish a universal design paradigm that democratizes and accelerates the development of mRNA medicines, unlocking a new era in precision biotechnology.