Topology- and Chirality-Aware SAFT-P for Phase Behavior of Multivalent Protein Condensates

Biomolecular condensates can be formed with multiple types of biomacromolecules via liquid–liquid phase separation (LLPS). In general, there is a belief that multivalent, patchy macromolecules, where not only valence but also the spatial arrangement of interaction sites and molecular chirality govern composition and phase behavior. To incorporate such interactions and study phase transitions in multicomponent systems we developed SAFT-P, a lattice-based, topology- and chirality-aware perturbation theory that extends Statistical Associating Fluid Theory (SAFT). Benchmarking SAFT-P on few- to many-component patchy-particle mixtures, we find near-exact agreement with Monte Carlo simulations. Unlike conventional SAFT, which is insensitive to patch topology, SAFT-P discriminates between two-patch particles with different shapes and quantitatively reproduces their distinct condensation phase diagrams in the pure-component limit. We further show that SAFT-P captures chiral effects: in a quaternary model comprising solvent, a chiral selector, and the left- and right-handed enantiomers of a target species, the theory accurately recovers equilibrium compositions and phase boundaries associated with chiral separation. Together, these results establish SAFT-P as a predictive, computational framework for topologically complex and chiral systems.