Learning to Dock: Geometric Deep Learning for Predicting Supramolecular Host-Guest Complexes

TL;DR

DeepHostGuest is a geometric deep-learning framework that predicts supramolecular host-guest binding with high accuracy, generalizes across diverse systems, and facilitates rational design and high-throughput screening.

Contribution

The paper introduces DeepHostGuest, a transferable geometric deep-learning model that accurately predicts host-guest binding conformations and energies, surpassing classical methods and enabling rational supramolecular design.

Findings

Achieves RMSD ≤ 2 Å for 80.8% test cases

Generalizes to crystalline sponge systems and large molecules

Correlates DFT affinities with experimental data across 876 complexes

Abstract

Predicting non-covalent host-guest recognition remains challenging due to the complex interplay of electrostatics, dispersion, and steric effects, and the limited transferability of existing docking approaches to synthetic supramolecular systems. Here we present DeepHostGuest, a geometric deep-learning framework that learns generalizable recognition principles directly from experimentally resolved host-guest structures. Hosts are encoded as electrostatic surfaces and guests as molecular graphs, enabling transferable learning across diverse supramolecular systems. DeepHostGuest achieves high-accuracy predictions (RMSD $\leq 2$ Angstrom for 80.8% of test cases), substantially outperforming classical docking without case-specific tuning. Notably, the method generalizes beyond its training domain to crystalline sponge systems, accurately capturing the binding of large amphiphilic molecules within metal-organic cages. Beyond predicting binding conformations, the structures generated by DeepHostGuest serve as a reliable basis for accurate binding free-energy calculations. Density Functional Theory (DFT)-calculated affinities correlate well with experiment, enabling structure-property relationships across 876 host-guest complexes spanning 34 host families. Interpretable feature analysis reveals that binding strength arises from a cooperative interplay of host polarity, guest hydrophobicity, and geometric complementarity, with distinct design regimes across supramolecular classes. Together, these results establish data-driven molecular recognition as a practical route to predictive supramolecular design, enabling high-throughput virtual screening and rational optimization of functional host-guest systems.