Geometric Deep Learning for Molecular Crystal Structure Prediction
Open Access
- 13 April 2023
- journal article
- research article
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 19 (14) , 4743-4756
Abstract
We develop and test new machine learning strategies for accelerating molecular crystal structure ranking and crystal property prediction using tools from geometric deep learning on molecular graphs. Leveraging developments in graph-based learning and the availability of large molecular crystal data sets, we train models for density prediction and stability ranking which are accurate, fast to evaluate, and applicable to molecules of widely varying size and composition. Our density prediction model, MolXtalNet-D, achieves state-of-the-art performance, with lower than 2% mean absolute error on a large and diverse test data set. Our crystal ranking tool, MolXtalNet-S, correctly discriminates experimental samples from synthetically generated fakes and is further validated through analysis of the submissions to the Cambridge Structural Database Blind Tests 5 and 6. Our new tools are computationally cheap and flexible enough to be deployed within an existing crystal structure prediction pipeline both to reduce the search space and score/filter crystal structure candidates.Keywords
All Related Versions
Funding Information
- Division of Chemistry (CHE-1955381)
- Deutsche Forschungsgemeinschaft (428315600)
- Natural Sciences and Engineering Research Council of Canada (PDF-545751-2020)
- Division of Materials Research (DMR-2118890)
This publication has 28 references indexed in Scilit:
- Next-Generation Force Fields from Symmetry-Adapted Perturbation TheoryAnnual Review of Physical Chemistry, 2016
- The Cambridge Structural DatabaseActa crystallographica Section B, Structural science, crystal engineering and materials, 2016
- Crystal structure prediction via particle-swarm optimizationPhysical Review B, 2010
- Knowledge-based model of hydrogen-bonding propensity in organic crystalsActa crystallographica Section B, Structural science, crystal engineering and materials, 2007
- Development and testing of a general amber force fieldJournal of Computational Chemistry, 2004
- Derivation of a scoring function for crystal structure predictionActa Crystallographica Section A Foundations of Crystallography, 2001
- Crystal Structure Prediction and the Cambridge Structural DatabaseMolecular Crystals and Liquid Crystals, 2001
- Stochastic Tunneling Approach for Global Minimization of Complex Potential Energy LandscapesPhysical Review Letters, 1999
- Crystal Structure Prediction based on Statistical PotentialsJournal of Molecular Modeling, 1998
- IsoStar: A library of information about nonbonded interactionsJournal of Computer-Aided Molecular Design, 1997