PM6-ML: The Synergy of Semiempirical Quantum Chemistry and Machine Learning Transformed into a Practical Computational Method

2025/01/03 by Martin Nováček, Jan Řezáč · 2 voices · 2 citations
Materials Science · Computer Science · Biochemistry, Genetics and Molecular Biology · #Machine Learning in Materials Science #Computational Drug Discovery Methods #Protein Structure and Dynamics

paper · doi:10.1021/acs.jctc.4c01330

openalex publication_date 2025/01/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/17

Abstract

Machine learning (ML) methods offer a promising route to the construction of universal molecular potentials with high accuracy and low computational cost. It is becoming evident that integrating physical principles into these models, or utilizing them in a Δ-ML scheme, significantly enhances their robustness and transferability. This paper introduces PM6-ML, a Δ-ML method that synergizes the semiempirical quantum-mechanical (SQM) method PM6 with a state-of-the-art ML potential applied as a universal correction. The method demonstrates superior performance over standalone SQM and ML approaches and covers a broader chemical space than its predecessors. It is scalable to systems with thousands of atoms, which makes it applicable to large biomolecular systems. Extensive benchmarking confirms PM6-ML's accuracy and robustness. Its practical application is facilitated by a direct interface to MOPAC. The code and parameters are available at https://github.com/Honza-R/mopac-ml.

Citations

• Development and testing of a general amber force field
• The ORCA quantum chemistry program package
• Software update: The ORCA program system—Version 5.0

Cited by

• AIQM2: organic reaction simulations beyond DFT
• Predictive Free Energy Simulations Through Hierarchical Distillation of Quantum Hamiltonians
• Benchmark of Approximate Quantum Chemical and Machine Learning Potentials for Biochemical Proton Transfer Reactions
• The Enduring Relevance of Semiempirical Quantum Mechanics

Discussions

• #seminar Thank you to Professor Jan Řezáč @iocbprague.bsky.social‬ for sharing the progress his group is making toward combining #quantumchemistry ⚛️ and #machinelearning 🤖 to model protein-ligand i [bsky, 5 points, 0 comments]
• #research #machinelearning #compchem PM6-ML: The Synergy of Semiempirical Quantum Chemistry and Machine Learning Transformed into a Practical Computational Method (Nováček, Řezáč) - JCTC: doi.org/10. [bsky, 5 points, 0 comments]

Related

• PM6-ML: The Synergy of Semiempirical Quantum Chemistry and Machine Learning Transformed into a Practical Computational Method
• Machine Learning Potentials: A Roadmap Toward Next-Generation Biomolecular Simulations
• Learning Together: Towards foundational models for machine learning interatomic potentials with meta-learning
• Automated Machine Learning Pipeline: Large Language Models-Assisted Automated Dataset Generation for Training Machine-Learned Interatomic Potentials
• The evolution of machine learning potentials for molecules, reactions and materials