Search papers, labs, and topics across Lattice.
This study introduces PHASE, a framework that transforms atomistic conformational ensembles into explicit statistical models by utilizing local Hamiltonians to represent residue couplings. By applying PHASE to the adenosine A2A receptor, the researchers demonstrated that these Hamiltonians can accurately reproduce microstate statistics and capture correlations between non-directly coupled residues. The approach not only organizes sampled ensembles along the A2A activation landscape but also facilitates a novel all-atom backmapping process, paving the way for efficient quantum-annealing applications.
Local residue couplings can accurately capture complex protein conformational dynamics, revealing hidden correlations and enabling efficient sampling along activation landscapes.
Protein function is governed by conformational ensembles, which can be viewed as high-dimensional probability distributions over molecular conformations. Yet the statistical organization of these distributions is often represented only implicitly, either through collections of simulation trajectories or within high-capacity generative models. Here, we introduce PHASE (Protein Hamiltonians for Sampling of Ensembles), a system-specific framework that converts atomistic conformational ensembles into an explicit and interpretable statistical model. Applied to ten conformational ensembles derived from approximately 37$\mu$s of atomistic simulations of the adenosine A2A receptor, Hamiltonians containing only local residue couplings within 6$\mathring{A}$ reproduce residue-wise and pairwise microstate statistics, including correlations between residues that are not directly coupled in the model. Moreover, independently fitted inactive and active reference Hamiltonians define an endpoint preference coordinate that organizes newly sampled ligand-, effector- and conformation-dependent ensembles along the A2A activation landscape without receiving these biochemical labels as model inputs. Finally, a cluster-conditioned all-atom reconstruction model preserves the prescribed residue microstate patterns of newly sampled configurations, closing the coarse-graining-sampling-backmapping cycle. The resulting discrete representation additionally admits direct QUBO encoding, enabling classical annealing and providing a route toward future quantum-annealing implementations. PHASE therefore provides a protein-general procedure for constructing compact, interpretable and atomistically realizable statistical models of protein conformational ensembles.
