Simulating Biomolecular Condensates Using Patchy Colloids

Proteins are ubiquitous in all forms of life, and in recent years the paradigm of structural biology has shifted from directly trying to find correspondences between a protein’s structure and its function, to taking into account also the context in which it is found in the cell. One of the emergent phenomena related to this observation is Liquid-Liquid Phase Separation (LLPS), in which a protein-rich and a solvent-rich domain are formed under specific conditions found in vitro that mimic the main factors of what is found in vivo, such as temperature, pH or ionic strength variations, introduction of molecular crowding agents, or the presence of scaffold proteins which sequester clients into pre-formed droplets. This field of study is of special interest due to the apparent ubiquity of this phenomenon with a wide range of proteins and partners, and due to its biological relevance for, for example, colocalization of enzymes for catalyzing specific reactions in the cell. This problem can be modeled as chemical reactions of transient oligomerization which, for longer timescales, leads to the formation of protein-rich and solvent-rich domains. Since this process involves large numbers of molecules in long timescales and wide spatial ranges, involving mostly weak and transient interactions, computational studies on LLPS demand highly coarse-grained molecular simulations to probe what happens in the mesoscale. With this in mind, we have developed a flexible simulation tool for Kinetic Monte Carlo simulations of colloids employing the Kern-Frenkel model of patchy particles to assess protein interactions in general, ranging from phenomena such as fiber formation to LLPS with crowding agents. The user can define an arbitrary number of particle types, each with specific interaction parameters and sites using square well potentials, and obtain information on the dynamics of assembly and disassembly of protein clusters and condensates. The program also allows for the calculation of observables such as the structure factor, fractal dimension of aggregates and kinetic constants for comparison with experiments.

This work was supported by FAPESP (2025/10983-9).