Reflection-Equivariant Diffusion for 3D Structure Determination from Isotopologue Rotational Spectra in Natural Abundance
arXiv:2310.11609 · doi:10.1063/5.0196620
Abstract
Structure determination is necessary to identify unknown organic molecules, such as those in natural products, forensic samples, the interstellar medium, and laboratory syntheses. Rotational spectroscopy enables structure determination by providing accurate 3D information about small organic molecules via their moments of inertia. Using these moments, Kraitchman analysis determines isotopic substitution coordinates, which are the unsigned coordinates of all atoms with natural isotopic abundance, including carbon, nitrogen, and oxygen. While unsigned substitution coordinates can verify guesses of structures, the missing signs make it challenging to determine the actual structure from the substitution coordinates alone. To tackle this inverse problem, we develop KREED (Kraitchman REflection-Equivariant Diffusion), a generative diffusion model that infers a molecule's complete 3D structure from its molecular formula, moments of inertia, and unsigned substitution coordinates of heavy atoms. KREED's top-1 predictions identify the correct 3D structure with >98% accuracy on the QM9 and GEOM datasets when provided with substitution coordinates of all heavy atoms with natural isotopic abundance. When substitution coordinates are restricted to only a subset of carbons, accuracy is retained at 91% on QM9 and 32% on GEOM. On a test set of experimentally measured substitution coordinates gathered from the literature, KREED predicts the correct all-atom 3D structure in 25 of 33 cases, demonstrating experimental applicability for context-free 3D structure determination with rotational spectroscopy.
arXiv commentadded software citations
References in corpus (26)
- Adam: A Method for Stochastic Optimization
- SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
- Array Programming with NumPy
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Denoising Diffusion Probabilistic Models
- Diffusion Models Beat GANs on Image Synthesis
- Deep Unsupervised Learning using Nonequilibrium Thermodynamics
- Classifier-Free Diffusion Guidance
- Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
- DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking
- GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation
- Diffusion Posterior Sampling for General Noisy Inverse Problems
- Equivariant Diffusion for Molecule Generation in 3D
- E(n) Equivariant Graph Neural Networks
- Structure-based Drug Design with Equivariant Diffusion Models
- Torsional Diffusion for Molecular Conformer Generation
- Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning
- Geometric Latent Diffusion Models for 3D Molecule Generation
- On the adequacy of untuned warmup for adaptive optimization
- Automated, context-free assignment of asymmetric rotor microwave spectra
- Continuous diffusion for categorical data
- Frame Averaging for Invariant and Equivariant Network Design
- Provably Strict Generalisation Benefit for Equivariant Models
- MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation
- FAENet: Frame Averaging Equivariant GNN for Materials Modeling
- Expressive Sign Equivariant Networks for Spectral Geometric Learning