Context-Aware Hydrophobicity Modeling: HydroMap and FastHydroMap

Abstract

Hydrophobicity governs a vast range of phenomena, from protein-protein interactions to nanomaterial assembly, and can be rigorously quantified by the dewetting free energy (Fdewet) of a molecule or surface. However, hydrophobicity remains widely treated as an additive property of amino acid identity, obscuring the fact that waters response is a collective property of the surface, shaped by curvature, chemical patterning, and neighboring residues. Direct calculation of Fdewet via specialized molecular simulations captures this collective behavior but is prohibitively slow, leaving in place broadly-used, decades-old sequence-based hydropathy scales that neglect the physics of solvation. Here we show that residue-level Fdewet can be predicted from a compact set of local water features (water structural signatures and residue-water potential energy) extracted from a brief and inexpensive all-atom simulation. We embed this insight in two models: HydroMap, which predicts Fdewet directly from water features, and FastHydroMap, a computationally inexpensive graph neural network surrogate trained on HydroMap that requires no solvent simulation. HydroMap and FastHydroMap capture context-dependent hydrophobicity that classical, sequence-only hydropathy scales miss. We demonstrate this across three protein systems: on an -synuclein amyloid filament, strongly dewetting interfaces align with unassigned peptide densities, revealing hidden binding sites; in calmodulin, hydrophobicity redistributes upon Ca2+ binding; and for Protein G, time-resolved hydrophobicity changes track the folding trajectory. Together, these models make Fdewet a computationally inexpensive descriptor for proteins, membranes, and other surfaces, enabling rapid scoring for materials design and a time-resolved view of dynamic hydrophobic-mediated processes such as protein folding. Significance StatementHydrophobicity, the tendency of surfaces to expel water, drives how proteins fold and how molecules recognize one another. For decades, it has widely been treated as a fixed property of an amino acid or chemical group, but water actually responds to the collective shape and chemistry of a surface, such as that presented by a protein, not to its components in isolation. Measuring this collective response from molecular simulation is rigorous but prohibitively slow. We show that it can instead be inferred from a compact set of features describing water structure and interactions near a surface, and we use this insight to build models that predict hydrophobicity rapidly and at residue resolution, enabling practical, physically grounded design of hydrophobic-mediated interactions.