Bits2Bonds: AI system accelerates discovery of RNA delivery polymers

A research group led by Professor Olivia Merkel, Chair of Drug Delivery at LMU Munich have announced a new computational platform that could significantly speed up the development of RNA-based medicines. The study introduces Bits2Bonds, the first tool to combine molecular dynamics (MD) simulations with machine learning (ML) for the de novo design of polymeric carriers capable of transporting therapeutic RNA into cells.

The breakthrough forms part of Professor Merkel’s European Research Council (ERC) Consolidator Grant project, ‘RatInhalRNA’, which focuses on creating advanced RNA delivery technologies specifically tailored for pulmonary administration.

Overcoming barriers in RNA delivery research

Developing effective delivery vehicles for therapeutic RNA is one of the main challenges in modern drug design. Experimental screening of large polymer libraries is notoriously labour-intensive, expensive and slow. Meanwhile, earlier computational tools have been limited by scarce training data and the heavy processing power required for accurate simulations.

Bits2Bonds looks to solve these problems by integrating coarse-grained MD simulations – used to mimic crucial biological events such as siRNA binding and membrane interactions – with machine learning systems capable of predicting and optimising molecular structures. By combining these approaches, the platform can rapidly analyse thousands of potential polymer candidates virtually, narrowing the field long before laboratory testing starts.

A new era for high-throughput RNA carrier design

“Our work demonstrates for the first time that combining physics-based simulation with data-driven optimisation can efficiently guide the discovery of entirely new materials for RNA therapeutics,” Professor Merkel said. “This method paves the way for a more rational, high-throughput design of polymeric delivery systems, moving us closer to personalised RNA medicines.”

The hybrid approach gives researchers a powerful means of exploring unexplored chemical space, identifying materials that traditional screening methods might miss. As a result, Bits2Bonds could dramatically shorten the timeline for developing clinically viable nanocarriers, particularly those able to deliver small interfering RNA (siRNA), which requires stability and precision to function safely.

Promising experimental validation

To test the platform, the team synthesised several polymer candidates that Bits2Bonds predicted would exhibit strong RNA-binding and delivery capabilities. Lab experiments then confirmed a strong correlation between the simulated behaviour and actual biological performance, demonstrating reliability of the integrated modelling approach.

The researchers report that the system is deliberately modular, meaning it can be adapted to investigate a wider range of polymer classes and nucleic acid types. This includes emerging therapeutic technologies such as messenger RNA (mRNA) vaccines and CRISPR-based gene-editing systems.

A versatile tool for future RNA therapeutics

Bits2Bonds allows researchers to explore vast numbers of potential materials in silico, which is a huge step for the field of nucleic acid delivery. With RNA-based therapies continuing to play an expanding role in personalised and precision medicine, the ability to design safer, more effective carriers at speed will be crucial.

As research continues under the RatInhalRNA programme, Professor Merkel’s team expects the platform to support the next wave of RNA therapeutic innovation – potentially reshaping the way such medicines are designed, tested and ultimately delivered to patients.