PENSim: A Toolkit for PEN-DNA Systems

Abstract

The PEN-DNA toolbox has emerged as a versatile framework for implementing chemical reaction networks with DNA and enzymes, enabling applications ranging from molecular sensing to neural computation. A simulation tool, DACCAD, has provided a standardized approach for translating network schematics into dynamical models, facilitating in silico design and validation. However, the evolution of the PEN-DNA toolbox - particularly the transition from inhibitor-based regulation to drain-template-based inactivation - introduces mechanisms that are only partially captured by existing simulation frameworks, limiting the ability to reliably simulate and design recent PEN-DNA systems.
In this context, we introduce PENSim, a simulation package that aims to realign computational modeling with current experimental practices. By extending the repertoire of simulated reactions and incorporating thermodynamic dependencies into kinetic rate calculations via NUPACK, PENSim enables a more faithful representation of modern PEN-DNA systems. In particular, accounting for thermodynamic effects is essential for accurately modeling inactivation via drain templates.
PENSim is validated through qualitative reproduction of existing experimental results, including inactivation-driven bistability, microRNA detection circuits, and linear classifiers based on neural networks. By bridging legacy modeling approaches with recent experimental advances, PENSim provides a step toward more realistic and flexible in silico design of systems based on the PEN-DNA toolbox.