AtomMOF: All-Atom Flow Matching for MOF-Adsorbate Structure Prediction

TL;DR

AtomMOF is a novel all-atom flow-based model that accurately predicts 3D structures of MOFs and adsorbates, improving structural validity and sampling efficiency over existing methods.

Contribution

We introduce AtomMOF, a scalable all-atom flow model that directly maps molecular graphs to 3D structures, incorporating interatomic potentials for better accuracy.

Findings

Increases match rate by 35% on BW dataset

Reduces RMSD by 32.64% on BW dataset

More sample-efficient than grand canonical Monte Carlo

Abstract

Deep generative models have shown promise for modeling metal-organic frameworks (MOFs), but existing approaches (1) rely on coarse-grained representations that assume fixed bond lengths and angles, and (2) neglect the MOF-adsorbate interactions, which are critical for downstream applications. We introduce AtomMOF, a scalable flow-based model built on an all-atom Diffusion Transformer that maps 2D molecular graphs of building blocks and adsorbates directly to equilibrium 3D structures without imposing structural constraints. We further present scaling laws for porous crystal generation, indicating predictable performance gains with increased model capacity, and introduce Feynman-Kac steering guided by machine-learned interatomic potentials to improve geometric validity and sampling stability. On the (MOF-only) BW dataset, AtomMOF increases the match rate by 35.00% and reduces RMSD by 32.64%. On the ODAC25 dataset (MOF-adsorbate), AtomMOF is substantially more sample-efficient than grand canonical Monte Carlo in recovering adsorption configurations and can identify candidates with lower adsorption energies than the reference dataset. Code is available at https://github.com/nayoung10/AtomMOF.