DiffSpectra: Molecular Structure Elucidation from Spectra using Diffusion Models

2025-07-09

AI Summary

Molecular structure elucidation from spectroscopic data is a fundamental challenge in chemical identification and drug discovery; however, traditional expert-based approaches lack scalability, while existing machine learning methods are constrained by reliance on reference databases or autoregressive SMILES generation, limiting generalization to novel molecules and neglecting 3D geometric information. This paper introduces DiffSpectra—the first end-to-end generative framework jointly modeling multimodal spectra (e.g., NMR, IR, MS) and both 2D and 3D molecular structures. Its key contributions are: (1) an SE(3)-equivariant diffusion model for geometry-aware 3D molecular generation; (2) SpecFormer, a unified spectral encoder that jointly represents heterogeneous spectra and injects conditional information; and (3) a Diffusion Molecule Transformer as the core denoising backbone. On standard benchmarks, DiffSpectra achieves 16.01% top-1 and 96.86% top-20 accuracy, empirically validating the efficacy of 3D structural modeling, spectral pretraining, and multimodal conditional fusion.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Molecular structure elucidation from spectra is a foundational problem in chemistry, with profound implications for compound identification, synthesis, and drug development. Traditional methods rely heavily on expert interpretation and lack scalability. Pioneering machine learning methods have introduced retrieval-based strategies, but their reliance on finite libraries limits generalization to novel molecules. Generative models offer a promising alternative, yet most adopt autoregressive SMILES-based architectures that overlook 3D geometry and struggle to integrate diverse spectral modalities. In this work, we present DiffSpectra, a generative framework that directly infers both 2D and 3D molecular structures from multi-modal spectral data using diffusion models. DiffSpectra formulates structure elucidation as a conditional generation process. Its denoising network is parameterized by Diffusion Molecule Transformer, an SE(3)-equivariant architecture that integrates topological and geometric information. Conditioning is provided by SpecFormer, a transformer-based spectral encoder that captures intra- and inter-spectral dependencies from multi-modal spectra. Extensive experiments demonstrate that DiffSpectra achieves high accuracy in structure elucidation, recovering exact structures with 16.01% top-1 accuracy and 96.86% top-20 accuracy through sampling. The model benefits significantly from 3D geometric modeling, SpecFormer pre-training, and multi-modal conditioning. These results highlight the effectiveness of spectrum-conditioned diffusion modeling in addressing the challenge of molecular structure elucidation. To our knowledge, DiffSpectra is the first framework to unify multi-modal spectral reasoning and joint 2D/3D generative modeling for de novo molecular structure elucidation.

Problem

• Elucidating molecular structures from multi-modal spectral data
• Overcoming limitations of traditional and retrieval-based methods
• Integrating 2D/3D geometry with spectral information generation

Innovation

• Uses diffusion models for molecular structure elucidation
• Integrates 2D and 3D molecular structures
• Employs transformer-based spectral encoder