@unpublished{BushuievBushuievdeJongeetal.2025, author = {Bushuiev, Roman and Bushuiev, Anton and de Jonge, Niek F. and Young, Adamo and Kretschmer, Fleming and Samusevich, Raman and Heirman, Janne and Wang, Fei and Zhang, Luke and D{\"u}hrkop, Kai and Ludwig, Marcus and Haupt, Nils A. and Kalia, Apurva and Brungs, Corinna and Schmid, Robin and Greiner, Russell and Wang, Bo and Wishart, David S. and Liu, Li-Ping and Rousu, Juho and Bittremieux, Wout and R{\"o}st, Hannes and Mak, Tytus D. and Hassoun, Soha and Huber, Florian and van der Hooft, Justin J.J. and Stravs, Michael A. and B{\"o}cker, Sebastian and Sivic, Josef and Pluskal, Tom{\´a}š}, title = {MassSpecGym: A benchmark for the discovery and identification of molecules}, series = {arXiv}, journal = {arXiv}, edition = {v3}, publisher = {arXiv}, doi = {10.48550/arXiv.2410.23326}, pages = {49}, year = {2025}, abstract = {The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a result, the vast majority of acquired MS/MS spectra remain uninterpreted, thereby limiting our understanding of the underlying (bio)chemical processes. Despite decades of progress in machine learning applications for predicting molecular structures from MS/MS spectra, the development of new methods is severely hindered by the lack of standard datasets and evaluation protocols. To address this problem, we propose MassSpecGym -- the first comprehensive benchmark for the discovery and identification of molecules from MS/MS data. Our benchmark comprises the largest publicly available collection of high-quality labeled MS/MS spectra and defines three MS/MS annotation challenges: de novo molecular structure generation, molecule retrieval, and spectrum simulation. It includes new evaluation metrics and a generalization-demanding data split, therefore standardizing the MS/MS annotation tasks and rendering the problem accessible to the broad machine learning community. MassSpecGym is publicly available at this https URL [https://github.com/pluskal-lab/MassSpecGym].}, subject = {Maschinelles Lernen}, language = {en} }