Reflectin A1 is an intrinsically disordered protein whose environmentally responsive self-assembly offers a promising foundation for bio-inspired functional materials. A central challenge is to connect the sequence-specific molecular interactions of reflectin to its collective assembly behavior across experimentally relevant length and time scales. This work develops a multiscale computational framework that bridges molecular interactions, effective colloidal potentials, and phase behavior in reflectin A1.

A top-down approach based on small-angle X-ray scattering (SAXS) represents reflectin A1 as an effective colloidal particle governed by short-range attractive and long-range repulsive (SALR) interactions. The resulting model captures key assembly transitions, including dispersed monomers, finite equilibrium clusters with controlled sizes, and, under sufficiently strong attraction, liquid–liquid phase separation-like condensates. Variation in pH and salt concentration modulates the balance between short-range attraction and long-range electrostatic repulsion, thereby controlling cluster formation and the accessible phase behavior.

To establish the molecular basis of the colloidal SALR description, atomistic molecular dynamics simulations, sequence-specific coarse-grained modeling, and colloidal-scale analysis were integrated in a bottom-up framework. Atomistic simulations identified a distributed network of attractive interactions involving aromatic, cationic, and sulfur-containing residues. These interactions were incorporated into a residue-level coarse-grained force field that reproduced the SALR-type effective potential of full-length reflectin A1 without fitting to experimentally derived interaction potentials. The predicted pH-dependent potentials were consistent with SAXS-informed potentials, connecting experimental observations to sequence-resolved molecular mechanisms. Simulations of a reduced reflectin-like L–R–L construct further indicate that attractive contacts are distributed across multiple sequence regions rather than concentrated within a single binding interface.

Application of the force field to simplified reflectin-inspired sequences identifies sequence features that retain pH-responsive SALR assembly. Preserving alternating attractive R domains and cationic L domains, together with the relative abundance of methionine and aromatic residues, produces gradual pH-dependent changes in effective interactions and more controlled cluster-size distributions. A minimal 61-residue reflectin-inspired sequence retains the characteristic SALR potential and reproduces the progression from monomeric states to finite clusters and phase-separated states.

These findings demonstrate that reflectin A1 self-assembly can be interpreted through an effective SALR interaction that emerges from transient, chemically heterogeneous, sequence-specific intermolecular contacts. The agreement between SAXS-informed colloidal modeling and bottom-up multiscale simulations provides a mechanistic basis for predicting and designing environmentally responsive intrinsically disordered protein materials with programmable self-assembly and phase behavior.