Abstract
Lipidic nanocarriers are a broad class of lipid-based vectors with proven potential for packaging and delivering emerging nucleic acid therapeutics. An important early step in the clinical development cycle is large-scale screening of diverse formulation libraries to assess particle quality and payload delivery efficiency. Due to the size of the screening space, this process can be both costly and time-consuming. To address this, computational models capable of predicting clinically relevant physio-chemical properties of dendrimer-lipid nanocarriers, along with their mRNA payload delivery efficiency in human cells are developed. The models are then deployed on a large theoretical nanocarrier pool consisting of over 4.5 million formulations. Top predictions are synthesised for validation using cell-based assays, leading to the discovery of a high quality, high performing, candidate. The methods reported here enable rapid, high-throughput, in silico pre-screening for high-quality candidates, and have great potential to reduce the cost and time required to bring mRNA therapies to the clinic.