Unraveling Molecular Structure: A Multimodal Spectroscopic Dataset for Chemistry

TL;DR

This paper introduces a large, multimodal spectroscopic dataset for 790,000 molecules, enabling machine learning models to integrate multiple spectroscopic techniques for improved molecular structure elucidation.

Contribution

The creation of a comprehensive multimodal spectroscopic dataset and benchmarks for machine learning tasks in molecular structure prediction.

Findings

Dataset includes simulated NMR, IR, and MS spectra for 790k molecules.

Benchmarks for structure elucidation, spectrum prediction, and functional group identification.

Potential to automate and improve molecular discovery processes.

Abstract

Spectroscopic techniques are essential tools for determining the structure of molecules. Different spectroscopic techniques, such as Nuclear magnetic resonance (NMR), Infrared spectroscopy, and Mass Spectrometry, provide insight into the molecular structure, including the presence or absence of functional groups. Chemists leverage the complementary nature of the different methods to their advantage. However, the lack of a comprehensive multimodal dataset, containing spectra from a variety of spectroscopic techniques, has limited machine-learning approaches mostly to single-modality tasks for predicting molecular structures from spectra. Here we introduce a dataset comprising simulated $^1$H-NMR, $^{13}$C-NMR, HSQC-NMR, Infrared, and Mass spectra (positive and negative ion modes) for 790k molecules extracted from chemical reactions in patent data. This dataset enables the development of foundation models for integrating information from multiple spectroscopic modalities, emulating the approach employed by human experts. Additionally, we provide benchmarks for evaluating single-modality tasks such as structure elucidation, predicting the spectra for a target molecule, and functional group predictions. This dataset has the potential automate structure elucidation, streamlining the molecular discovery pipeline from synthesis to structure determination. The dataset and code for the benchmarks can be found at https://rxn4chemistry.github.io/multimodal-spectroscopic-dataset.