Simulating Tandem Mass Spectra for Small Molecules using a General-Purpose Large-Language Model

Abstract We show a practical application of the Google Gemini large-language-model for simulating tandem mass spectra for compounds from the Blood Exposome Database. This approach bypasses the need for domain-specific model training, suggesting that the chemical fragmentation knowledge could be latently encoded within the Gemini model. General-purpose LLMs represent a useful and accessible tool for expanding in-silico spectral libraries and may accelerate the compound annotation in mass spectrometry-based metabolomics and exposomics. Competing Interest Statement The authors have declared no competing interest. Footnotes Funder Information Declared National Institute of Environmental Health Sciences, https://ror.org/00j4k1h63, U24ES035386, R24ES036917, R01ES035478, P30ES023515, R01ES032831, R01ES033688 National Center for Advancing Translational Sciences, UL1TR004419 Copyright The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license.