Year: 2026
Type: preprint
Abstract: We report a new methodology for the discovery of polymer-based gene vectors that enables the simultaneous optimization of polymer composition and polyplex formulation, while bypassing the need to prepare, isolate, and characterise new candidates using conventional polymer synthesis. At the core of this approach is the combination of computer-aided experimental design, in situ screening of polymer composition and polyplex formulation, and machine learning, which allows us to treat polymer synthesis and formulation as a single optimizable problem. Central to this methodology is a new reactive and water-soluble poly(acetylene), poly(N-(propargyl)-2-(aminooxy)-acetamide), which can be functionalized in aqueous conditions with readily available aldehydes. This provides candidate polymers under biologically compatible conditions, which are then directly complexed with nucleic acids and evaluated within the same workflow. Iterative modelling then correlates inputs readily controlled by the operator—such as reagent concentrations and ratios—with the key performance metric, transfection efficiency, guiding subsequent rounds of candidate generation. In doing so, we move beyond post hoc correlations of physicochemical properties to establish an iterative, data-driven methodology for materials discovery. Ultimately, this methodology has unlocked the potential of poly(acetylene)s, a polymer family not previously explored in this context, to move beyond compositional mimicry and capture the dynamic helicity that underpins nucleic acid complexation and delivery in natural systems.
Source: ChemRxiv
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
Awards: G1044684712, G1105844351, G1455481585, G1496055848, G1938719665, G355572793, G4092263876, G4462967385, G47840581, G491712853, G5260651846, G5278945566, G5404395142, G5404935707, G5635532810, G5701233768, G624877701, G6258374248, G6319620625, G67688110, G6904152975, G7805626310, G7863371703, G8497050386
OpenAlex ID: W7160182814