Topological potentials guiding protein self-assembly

Ivan Spirandelli, Arnur Nigmetov, Dmitriy Morozov, Myfanwy E. EvansarXiv 2025 · DOI · arXiv

Abstract

Simulating the spontaneous self-assembly of proteins into functional complexes is a major challenge in computational biology due to the rugged energy landscapes and the short-range nature of traditional physical potentials. This paper introduces a novel "topological potential" based on persistent homology that acts as a long-range guiding bias. By integrating this topological term with standard morphometric potentials, researchers achieved up to a 16-fold increase in simulation success rates, enabling the assembly of complex viral and human protein structures that were previously computationally inaccessible.

Introduction

In the realm of molecular simulation, predicting how protein subunits click together—a process known as self-assembly—is often a "needle in a haystack" problem. The primary obstacle is kinetic trapping. Most high-fidelity models rely on short-range potentials, such as solvation-free energy and van der Waals forces, which only become effective when subunits are in close proximity. Consequently, the energy landscape is filled with non-functional local minima where proteins get stuck in incorrect orientations. Without a long-range force to guide them, subunits perform a largely random search, often failing to find the native state within feasible timescales.

Methodology: The Hybrid Potential

The authors propose a hybrid energy function, $E_{comb}$, that balances local physical realism with global topological guidance:

  1. The Morphometric Potential ($F^*_{sol}$): This standard component calculates solvation-free energy based on four geometric measures: volume, surface area, and two types of curvature. While accurate at short ranges, it is inherently limited for distant interactions.
  2. The Topological Potential ($\mathcal{T}$): This novel contribution utilizes Topological Data Analysis (TDA), specifically persistent homology, to quantify the system's shape across multiple scales. It calculates "weighted total persistence"—the sum of the lifetimes of topological features like connected components ($P_0$), loops ($P_1$), and voids ($P_2$).
Topological features such as loops and voids are quantified using persistent homology to create a guiding potential.

By mixing these terms ($E_{comb} = \mu F^*_{sol} + (1 - \mu)\mathcal{T}$), the topological potential creates a "guiding funnel" that pulls subunits together and aligns them correctly from a distance, while the morphometric term handles the final, precise docking.

Key Insights and Results

The study evaluated this method using structural data from the Protein Data Bank (PDB) for systems including the Tobacco Mosaic Virus (TMV), Hepatitis B core protein, and SARS-CoV-2 ORF9b.

A complex protein assembly successfully guided by the topological potential.

The "Funnel" Effect

The authors found that $P_0$ (zero-dimensional persistence) acts as a long-range attraction, pulling subunits together. Meanwhile, $P_1$ and $P_2$ provide the orientational cues needed to avoid clashing or misaligning. A rigorous mathematical link was established between these persistence measures and the Euler characteristic, providing a theoretical justification for topology as a proxy for physical shape.

Performance Gains

The inclusion of the topological potential led to dramatic improvements in success rates:

  • TMV Dimerization: Success rates jumped from 5% to 85%—a 16-fold increase.
  • Hepatitis B Core Protein: This system failed to assemble at all (0% success) using the standard model but achieved a 77.5% success rate with the topological bias.
  • SARS-CoV-2 ORF9b: Success increased from 2.5% to 42.5%.

Conclusion and Future Directions

Topological potentials represent a significant shift in simulation strategy, moving TDA from a post-hoc analysis tool to an active guiding force. While current models treat proteins as rigid bodies, future work aims to incorporate protein flexibility. This approach holds immense promise for drug design and the development of novel nanomaterials, where navigating complex assembly pathways is the primary bottleneck.

Summary updated

Abstract

The simulated self-assembly of molecular building blocks into functional complexes is a key area of study in computational biology and materials science. Self-assembly simulations of proteins using physically-motivated potentials for non-polar interactions, can identify the biologically correct assembly as the energy-minimizing state. Short-range potentials, however, produce rugged energy landscapes, which lead to simulations becoming trapped in non-functional local minimizers. Successful self-assembly simulations depend on the physical realism of the driving potentials as well as their ability to efficiently explore the configuration space. We introduce a long-range topological potential, quantified via weighted total persistence, and combine it with the morphometric approach to solvation-free energy. This combination improves the assembly success rate in simulations of the tobacco mosaic virus dimer and other protein complexes by up to sixteen-fold compared with the morphometric model alone. It further enables successful simulation in systems that don't otherwise assemble during the examined timescales. Compared to previous topology-based work, which has been primarily descriptive, our approach uses topological measures as an active energetic bias that is independent of electrostatics or chemical specificity and depends only on atomic coordinates. Therefore, the method can, in principle, be applied to arbitrary systems where such coordinates are optimized. Integrating topological descriptions into an energy function offers a general strategy for overcoming kinetic barriers in molecular simulations, with potential applications in drug design, materials development, and the study of complex self-assembly processes.
BibTeX
@misc{spirandelli2025topological,
  author = {Ivan Spirandelli and Arnur Nigmetov and Dmitriy Morozov and Myfanwy E. Evans},
  title = {{Topological potentials guiding protein self-assembly}},
  year = {2025},
  eprint = {2508.15321},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2508.15321},
}
Figure