Enhancing NMR Shielding Predictions of Atoms-in-Molecules Machine Learning Models with Neighborhood-Informed Representations

2025/10/07 by Surajit Das, Das, Surajit, Raghunathan Ramakrishnan +1 · 1 voice
Chemistry · Materials Science · Physics and Astronomy · #Advanced NMR Techniques and Applications #Machine Learning in Materials Science #X-ray Diffraction in Crystallography #physics.chem-ph

paper · pdf · doi:10.48550/arxiv.2510.05623

arxiv published 2025/10/07 · arxiv updated 2025/12/11

Abstract

Accurate prediction of nuclear magnetic resonance (NMR) shielding with machine learning (ML) models remains a central challenge for data-driven spectroscopy. We present atomic variants of the Coulomb matrix (aCM) and bag-of-bonds (aBoB) descriptors, and extend them using radial basis functions (RBFs) to yield smooth, per-atom representations (aCM-RBF, aBoB-RBF). Local structural information is incorporated by augmenting each atomic descriptor with contributions from the n nearest neighbors, resulting in the family of descriptors, aCM-RBF(n) and aBoB-RBF(n). For 13C shielding prediction on the QM9NMR dataset (831,925 shielding values across 130,831 molecules), aBoB-RBF(4) achieves an out-of-sample mean error of 1.69 ppm, outperforming models reported in previous studies. While explicit three-body descriptors further reduce errors at a higher cost, aBoB-RBF(4) offers the best balance of accuracy and efficiency. Benchmarking on external datasets comprising larger molecules (GDBm, Drug12/Drug40, and pyrimidinone derivatives) confirms the robustness and transferability of aBoB-RBF(4), establishing it as a practical tool for ML-based NMR shielding prediction.

Citations

• Chemical Space-Informed Machine Learning Models for Rapid Predictions of X-ray Photoelectron Spectra of Organic Molecules
• Universal neural network potentials as descriptors: Towards scalable chemical property prediction using quantum and classical computers
• MACE-OFF: Transferable Short Range Machine Learning Force Fields for Organic Molecules
• MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields
• Revving up 13C NMR shielding predictions across chemical space:\n Benchmarks for atoms-in-molecules kernel machine learning with new data for\n 134 kilo molecules
• IMPRESSION -- Prediction of NMR Parameters for 3-dimensional chemical\n structures using Machine Learning with near quantum chemical accuracy
• SchNet – A deep learning architecture for molecules and materials
• Localized Coulomb Descriptors for the Gaussian Approximation Potential
• Many Molecular Properties from One Kernel in Chemical Space
• Fourier series of atomic radial distribution functions: A molecular fingerprint for machine learning models of quantum chemical properties
• MOPAC: A semiempirical molecular orbital program
• The Graph Neural Network Model

Discussions

• I have presented my Master's thesis work ("Enhancing NMR Shielding Predictions of Atoms-in-Molecules Machine Learning Models with Neighborhood-Informed Representations") at the In-house Symposium orga [bsky, 3 points, 0 comments]

Related

• Revving up 13C NMR shielding predictions across chemical space:\n Benchmarks for atoms-in-molecules kernel machine learning with new data for\n 134 kilo molecules
• Highly Accurate Prediction of NMR Chemical Shifts from Low-Level Quantum Mechanics Calculations Using Machine Learning
• A deep learning model for chemical shieldings in molecular organic solids including anisotropy
• Efficient Prediction of Transition-Metal NMR Chemical Shifts Using Machine Learning: Do Two-Dimensional Descriptors Suffice?
• Towards a Unified Benchmark and Framework for Deep Learning-Based Prediction of Nuclear Magnetic Resonance Chemical Shifts