Structural Nucleic Acid Technology
Structural Nucleic Acid Technology
Embedded Files
Nucleic acid structures are fascinating materials shaped by evolutionary processes.
We aim to engineer various types of nucleic acid systems with programmable structures and artificial functionalities.
By combining modeling, design, and experiments, we seek to develop novel nucleic acid architectures and bring new concepts to the bio-nano systems.
Self-assembly principles
Self-assembly describes how molecular components autonomously organize into ordered structures through local interactions.
For example, DNA origami offers a model system for investigating how programmable molecular interactions drive the formation of nanoscale structures.
By tuning sequence design and assembly conditions, the formation and structural fidelity can be systematically explored and experimentally validated.
In silico design and prediction
The computational framework predicts the shape and mechanical behavior of DNA assemblies from their connectivity and sequence information.
By combining structural and electrostatic elements, the finite element-based model estimates the equilibrium configurations and mechanical responses of DNA assemblies.
This enables in silico evaluation and refinement before experimental synthesis, supporting the rational design of complex DNA architectures.
DNA assembly dynamics
Structured DNA assemblies can undergo thermal fluctuations and dynamic reconfiguration in response to environmental changes.
This computational framework combines structural and hydrodynamic finite element models within Langevin dynamics simulations.
The structural model describes mechanical and electrostatic interactions, while the hydrodynamic model captures solvent viscosity and random thermal forces.
By integrating finite element vectors and matrices, the framework predicts equilibrium dynamics, molecular fluctuations, and time-dependent conformational changes of DNA assemblies.
Chemo-mechanical modeling
DNA-binding molecules can change the shape and mechanical properties of structured DNA assemblies by modifying local DNA geometry and stiffness.
The chemo-mechanical framework combines molecular-level properties, binding models, and structural analysis.
This approach provides a route to designing DNA assemblies with chemically tunable shapes and mechanical responses, such as changes in global deformation and supercoiling instability.
Learning DNA assembly shapes
DNA assemblies can be viewed as networks of molecular connections defined by sequence and topology.
Graph neural networks, combined with mechanical modeling, learn how these connectivity patterns determine the final shape of DNA structures.
The fast prediction capability enables inverse design of DNA assemblies with target nanoscale geometries.
Computational Mechanics
Computational Mechanics
Computational approaches rooted in mechanics and physics enable the simulation of complex systems and the prediction of previously unexplored phenomena.
Developing fundamental algorithms and integrating diverse approaches, such as atomic-scale methods, continuum-scale techniques, and neural networks, remain exciting challenges.
Our focus is on establishing and refining mathematical descriptions of intriguing physical systems to create state-of-the-art computational models.
Time integration
Time integration methods for structural dynamics are developed, aiming at reliable algorithms for large-scale simulations.
One focus is on explicit schemes that offer improved accuracy, efficiency, or stability over conventional methods, for problems ranging from micro to macro scales.
Ongoing work extends these methods toward higher-order accuracy, nonlinear and time-varying systems, and computationally efficient implementations.
Page updated
Report abuse