Maria Julia Maristany

Biomolecular Condensates Through a Computational Microscope

Keywords: Phase separation, Biophysics, Molecular Dynamics.

Biomolecular condensates are membrane-less compartments inside cells that are essential in organising and controlling various cellular processes. In this thesis, I investigate how these structures form and behave using computer simulations and biophysical methods. I focus on how the properties and interactions between single molecules influence when and how these compartments form, how stable they are, and how they organise on a large scale. My research shows that simplified, coarse-grained computer models can effectively capture the complex behaviours of these systems, revealing how basic forces like electrical charges and specific molecular attractions can drive the formation of condensates in different biological contexts, in a predictable and tuneable manner. 

One of my key findings is that molecular valency — essentially the number of binding sites available on a molecule — strongly controls the formation of these condensates and affects their material properties. By looking at prion-like proteins and other functional molecular constructs, I show how specific building blocks impact condensate behaviour by changing how molecules interact with each other. This concept also holds for certain types of protein-based structures and chromatin, the substance that organises our DNA, linking microscopic interactions to larger structural effects. 

In this work, I also explore in high resolution the high-order organization of chromatin, the structure responsible for packaging our DNA, revealing how specific single-molecule properties govern its behaviour and assembly. My findings show that chromatin condensation is strongly influenced by the flexibility and length of linker DNA and the arrangement of nucleosomes, which can be modulated to fine-tune chromatin stability. Studying chromatin is uniquely challenging due to its highly dynamic and multiscale nature—it needs to be compact enough to fit within the cell nucleus, yet flexible and accessible for processes like gene expression and DNA repair. Traditional experimental approaches often lack the spatial and temporal resolution to capture chromatin’s real-time, large-scale interactions, which vary from atomic-level interactions to whole-chromosome structures. To address these limitations, I use multiscale computational models that allow me to simulate chromatin behaviour across different scales, providing insights into the fundamental principles driving chromatin organization and stability that are difficult to achieve through experimental methods alone. 

This work underlines how important molecules' shape and flexibility are for forming and maintaining these cellular compartments. It demonstrates that a combined approach—considering molecular interactions, structure, and multiscale modelling—is necessary to understand the principles that control these processes. I argue that to truly capture the complex behaviour of these condensates, we need to integrate experimental data with advanced simulations, moving beyond simple models. My research provides a toolkit for doing so, by increasing the resolution of traditional experimental imaging techniques with the aid of a computational microscope, and by providing, via optimized algorithms, the blueprint for designing biomolecular condensates with specific properties, offering new insights into how cells are organized and potential applications for diseases where these processes go awry.

How the work contributes to bridge fields of biology, physics and/or mathematics:

My work intersects biology, physics, and computational chemistry. I employ physics-based computational models to understand complex biological systems, like molecular condensates, exploring how molecular interactions drive organization and phase behaviour, and providing a multiscale perspective to explain condensates’ biological functionality rooted in first principle physics. My findings reveal how different physical properties of single biological molecules, like valency and flexibility, regulate their condensation and biological function. Collaborations with groups in Cambridge, UT Southwestern, and MBL ensure predictions are backed by in vitro and in vivo experiments. My computational techniques then enhance experimental resolution and link molecular to mesoscale phenomena.

Department of Physics, University of Cambridge

Rosana Collepardo-Guevara Lab