NMR-Solver: automated structure elucidation via large-scale spectral matching and physics-guided fragment...

Area of Science:

• Organic Chemistry
• Computational Chemistry
• Spectroscopy

Background:

• Nuclear Magnetic Resonance (NMR) spectroscopy is crucial for determining molecular structures.
• Interpreting NMR spectra is complex, time-consuming, and requires expertise, especially for novel compounds.
• Existing automated methods often lack accuracy and robustness in real-world applications.

Purpose of the Study:

• To develop a practical and interpretable framework for automated molecular structure elucidation using NMR spectra.
• To address the limitations of current methods in accuracy and applicability.
• To provide a scalable and chemically meaningful solution for structure determination.

Main Methods:

• Developed NMR-Solver, a framework integrating large-scale spectral matching with physics-guided molecular optimization.
• Exploited atomic-level structure-spectrum relationships in NMR data.
• Utilized computational NMR analysis, deep learning, and interpretable chemical reasoning.

Main Results:

• NMR-Solver demonstrated strong generalization and robustness across simulated, literature, and real-world experimental data.
• The framework provides practical utility for determining small organic molecule structures from 1H and 13C NMR spectra.
• Achieved accurate and interpretable molecular structure elucidation.

Conclusions:

• NMR-Solver offers a significant advancement in automated molecular structure elucidation.
• The integrated approach establishes a generalizable paradigm for solving inverse problems in molecular science.
• Facilitates scalable, automated, and chemically meaningful structure determination.