Forschung und Projekte

Geometry-based simulation of self assembly

We develop geometric approaches to understand how solvent-mediated interactions shape the structure and assembly of complex systems. A central methodological theme is the use of coarse-grained models combined with morphometric descriptions of solvation, allowing solvent effects to be captured through geometric measures rather than detailed microscopic simulations. We analyse how entropic forces arising from the surrounding liquid environment constrain the conformations of flexible biopolymers and drive the organisation of particulate systems. By focusing on geometry, we reveal how solvent structure can stabilize helical folds, influence polymer topology, and promote nontrivial collective assembly. This unified methodology highlights how solvent-induced entropic interactions can be systematically described through geometric principles, providing a transferable framework for studying self-organization across biological and soft-matter systems.

Related article: Complex forms from simple building blocks

Related papers

Topological potentials guiding protein self-assembly

Can solvents tie knots? Helical folds of biopolymers in liquid environments

Solvation, geometry, and assembly of the tobacco mosaic virus

Exotic self-assembly of hard spheres in a morphometric solvent

Periodic entanglements

We study the many facets of periodic entanglements found in various biological, molecular, and chemical structures like polymers, liquid crystals, and DNA origami. We use techniques from geometry, topology, combinatorics and graph theory to enumerate and characterise potential structures, in many cases using periodic graphs or triply-periodic minimal surfaces as scaffolds for the structures. In this case, tangling, graphs and surfaces are all related objects of study. On the other hand, we are also developing techniques for the characterisation of these structures, where we look at crossing diagrams and related invariants.

Related article: Complexity explained: Symmetric Tangling of Honeycomb Networks

Related papers

Measuring the entanglement complexity of 3-periodic networks through the untangling number

The untangling number of 3-periodic tangles

Diagrammatic representations of 3-periodic entanglements

Ideal geometry of periodic entanglements

Periodic entanglement III: tangled degree-3 finite and layer net intergrowths from rare forests

Periodic entanglement II: weavings from hyperbolic line patterns

Periodic entanglement I: nets from hyperbolic reticulations

Topological Data Analysis, Theory and Applications

In this group, we work on both aspects of topological data analysis. On the theoretical side, we develop tools in computational geometry, such as  stability results for geometric structures and new ways to measure distances between datasets. On the applied side, we apply TDA to complicated materials  and systems. For example, we have worked with fibrous materials and entangled pancreatic networks. We also use TDA to guide the assembly of proteins and to study dynamic structure characterisation. We are interested in expanding these methods to other complex systems.

Related papers

Topological Analysis of Multi-Network Threading in the Pancreas

Topological potentials guiding protein self-assembly

The medial axis of any closed bounded set is locally Lipschitz stable with respect to the Hausdorff distance under ambient diffeomorphisms

Bregman–Hausdorff Divergence: Strengthening the Connections Between Computational Geometry and Machine Learning

Cocoon microstructures through the lens of topological persistence

Framework materials and tensegrity structures

Framework materials can be understood as an embedded graph with edge-length constraints. Introducing an energy functional by adding contracting (cables) and expansive edges (struts) to this framework makes it a tensegrity (tensile integrity). This physical system can be modeled as a polynomial optimization problem, making related numerical strategies feasible.

We can tackle the modeling and equilibration of complicated real-world structures such as cylinder packings by developing a robust and general-purpose Riemannian optimization package based on homotopy continuation in combination with a geometric model for the contact between two filaments in tight contact based on tensegrities. This approach allows us to explore these structures’ deformative mechanisms, occasionally revealing an unexpected dilatant property known as auxeticity.

Related papers

Computing Euclidean distance and maximum likelihood retraction maps for constrained optimization

Robust geometric modeling of 3-periodic tensegrity frameworks using Riemannian optimization

Reentrant tensegrity: A three-periodic, chiral, tensegrity structure that is auxetic

Geometric auxetics