DiffMS: Diffusion Generation of Molecules Conditioned on Mass Spectra
This work addresses the problem of de novo molecular structure elucidation from mass spectra, which is crucial for accelerating scientific discovery in chemistry and related fields.
This paper introduces DiffMS, a formula-restricted encoder-decoder generative network that achieves state-of-the-art performance in generating molecular structures conditioned on mass spectra. The method utilizes a transformer encoder for mass spectra and a discrete graph diffusion decoder, which is pre-trained on a large dataset of fingerprint-structure pairs.
Mass spectrometry plays a fundamental role in elucidating the structures of unknown molecules and subsequent scientific discoveries. One formulation of the structure elucidation task is the conditional de novo generation of molecular structure given a mass spectrum. Toward a more accurate and efficient scientific discovery pipeline for small molecules, we present DiffMS, a formula-restricted encoder-decoder generative network that achieves state-of-the-art performance on this task. The encoder utilizes a transformer architecture and models mass spectra domain knowledge such as peak formulae and neutral losses, and the decoder is a discrete graph diffusion model restricted by the heavy-atom composition of a known chemical formula. To develop a robust decoder that bridges latent embeddings and molecular structures, we pretrain the diffusion decoder with fingerprint-structure pairs, which are available in virtually infinite quantities, compared to structure-spectrum pairs that number in the tens of thousands. Extensive experiments on established benchmarks show that DiffMS outperforms existing models on de novo molecule generation. We provide several ablations to demonstrate the effectiveness of our diffusion and pretraining approaches and show consistent performance scaling with increasing pretraining dataset size. DiffMS code is publicly available at https://github.com/coleygroup/DiffMS.