{"url":"/dataset/massspecgym","name":"MassSpecGym","full_name":"MassSpecGym: A benchmark for the discovery and identification of molecules","description_markdown":"MassSpecGym provides three challenges for benchmarking the discovery and identification of new molecules from MS/MS spectra:\r\n\r\n- 💥 ***De novo* molecule generation** (MS/MS spectrum → molecular structure)\r\n    - ✨ **Bonus chemical formulae challenge** (MS/MS spectrum + chemical formula → molecular structure)\r\n- 💥 **Molecule retrieval** (MS/MS spectrum → ranked list of candidate molecular structures)\r\n    - ✨ **Bonus chemical formulae challenge** (MS/MS spectrum → ranked list of candidate molecular structures with ground-truth chemical formulae)\r\n- 💥 **Spectrum simulation** (molecular structure → MS/MS spectrum)\r\n    - ✨ **Bonus chemical formulae challenge** (molecular structure → MS/MS spectrum; evaluated on the retrieval of molecular structures with ground-truth chemical formulae)\r\n\r\nThe provided challenges abstract the process of scientific discovery from biological and environmental samples into well-defined machine learning problems with pre-defined datasets, data splits, and evaluation metrics.","description_withheld":null,"homepage":"https://github.com/pluskal-lab/MassSpecGym","introduced_date":"2024-10-30","introduced_date_note":null,"introduced_by":{"paper":"/paper/massspecgym-a-benchmark-for-the-discovery-and","title":"MassSpecGym: A benchmark for the discovery and identification of molecules","first_author":"Roman Bushuiev","url":null},"license":{"name":"MIT","url":null},"modalities":[{"name":"Biology","url":"/datasets/modality/biology"}],"tasks":[{"name":"De novo molecule generation from MS/MS spectrum","url":"/task/de-novo-molecule-generation-from-ms-ms","datasets_with_task":"/datasets/task/de-novo-molecule-generation-from-ms-ms"},{"name":"Molecule retrieval from MS/MS spectrum","url":"/task/molecule-retrieval-from-ms-ms-spectrum","datasets_with_task":"/datasets/task/molecule-retrieval-from-ms-ms-spectrum"},{"name":"MS/MS spectrum simulation","url":"/task/ms-ms-spectrum-simulation","datasets_with_task":"/datasets/task/ms-ms-spectrum-simulation"},{"name":"De novo molecule generation from MS/MS spectrum (bonus chemical formulae)","url":"/task/de-novo-molecule-generation-from-ms-ms-1","datasets_with_task":"/datasets/task/de-novo-molecule-generation-from-ms-ms-1"},{"name":"MS/MS spectrum simulation (bonus chemical formulae)","url":"/task/ms-ms-spectrum-simulation-bonus-chemical","datasets_with_task":"/datasets/task/ms-ms-spectrum-simulation-bonus-chemical"},{"name":"Molecule retrieval from MS/MS spectrum (bonus chemical formulae)","url":"/task/molecule-retrieval-from-ms-ms-spectrum-bonus","datasets_with_task":"/datasets/task/molecule-retrieval-from-ms-ms-spectrum-bonus"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["MassSpecGym"],"data_loaders":[{"repo":"https://github.com/pluskal-lab/massspecgym","url":"https://github.com/pluskal-lab/massspecgym","frameworks":["pytorch"]}],"num_papers_in_archive":4,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/de-novo-molecule-generation-from-ms-ms-1","task":"De novo molecule generation from MS/MS spectrum (bonus chemical formulae)","dataset_variant":"MassSpecGym","rows":9,"metrics":["Top-1 Accuracy","Top-1 MCES","Top-1 Tanimoto","Top-10 Accuracy","Top-10 MCES","Top-10 Tanimoto"],"first_row_in_archive_order":{"model":"DiffMS","paper":"/paper/diffms-diffusion-generation-of-molecules","metrics":{"Top-1 Accuracy":"2.3","Top-1 MCES":"18.45","Top-1 Tanimoto":"0.28","Top-10 Accuracy":"4.25","Top-10 MCES":"14.73","Top-10 Tanimoto":"0.39"},"code_links":[{"title":"coleygroup/diffms","url":"https://github.com/coleygroup/diffms"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/molecule-retrieval-from-ms-ms-spectrum-bonus","task":"Molecule retrieval from MS/MS spectrum (bonus chemical formulae)","dataset_variant":"MassSpecGym","rows":8,"metrics":["Hit rate @ 1","Hit rate @ 5","Hit rate @ 20","MCES @ 1"],"first_row_in_archive_order":{"model":"JESTR","paper":"/paper/jestr-joint-embedding-space-technique-for","metrics":{"Hit rate @ 1":"11.85","Hit rate @ 20":"61.46","Hit rate @ 5":"32.95"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/molecule-retrieval-from-ms-ms-spectrum-on","task":"Molecule retrieval from MS/MS spectrum","dataset_variant":"MassSpecGym","rows":8,"metrics":["Hit rate @ 1","Hit rate @ 5","Hit rate @ 20","MCES @ 1"],"first_row_in_archive_order":{"model":"JESTR_NR","paper":"/paper/jestr-joint-embedding-space-technique-for","metrics":{"Hit rate @ 1":"15.62","Hit rate @ 20":"60.55","Hit rate @ 5":"37.47"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/ms-ms-spectrum-simulation-bonus-chemical","task":"MS/MS spectrum simulation (bonus chemical formulae)","dataset_variant":"MassSpecGym","rows":4,"metrics":["Hit Rate @ 1","Hit Rate @ 5","Hit Rate @ 20"],"first_row_in_archive_order":{"model":"FraGNNet","paper":"/paper/massspecgym-a-benchmark-for-the-discovery-and","metrics":{"Hit Rate @ 1":"31.93","Hit Rate @ 20":"82.70","Hit Rate @ 5":"63.20"},"code_links":[{"title":"pluskal-lab/massspecgym","url":"https://github.com/pluskal-lab/massspecgym"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/ms-ms-spectrum-simulation-on-massspecgym","task":"MS/MS spectrum simulation","dataset_variant":"MassSpecGym","rows":4,"metrics":["Cosine Similarity","Jensen-Shannon Similarity","Hit Rate @ 1","Hit Rate @ 5","Hit Rate @ 20"],"first_row_in_archive_order":{"model":"FraGNNet","paper":"/paper/massspecgym-a-benchmark-for-the-discovery-and","metrics":{"Cosine Similarity":"0.52","Hit Rate @ 1":"46.64","Hit Rate @ 20":"83.58","Hit Rate @ 5":"72.56","Jensen-Shannon Similarity":"0.47"},"code_links":[{"title":"pluskal-lab/massspecgym","url":"https://github.com/pluskal-lab/massspecgym"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/de-novo-molecule-generation-from-ms-ms","task":"De novo molecule generation from MS/MS spectrum","dataset_variant":"MassSpecGym","rows":3,"metrics":["Top-1 Accuracy","Top-1 MCES","Top-1 Tanimoto","Top-10 Accuracy","Top-10 MCES","Top-10 Tanimoto"],"first_row_in_archive_order":{"model":"Random chemical generation","paper":"/paper/massspecgym-a-benchmark-for-the-discovery-and","metrics":{"Top-1 Accuracy":"0.00","Top-1 MCES":"28.59","Top-1 Tanimoto":"0.07","Top-10 Accuracy":"0.00","Top-10 MCES":"25.72","Top-10 Tanimoto":"0.10"},"code_links":[{"title":"pluskal-lab/massspecgym","url":"https://github.com/pluskal-lab/massspecgym"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/diffms-diffusion-generation-of-molecules","title":"DiffMS: Diffusion Generation of Molecules Conditioned on Mass Spectra","date":"2025-02-13","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/madgen-mass-spec-attends-to-de-novo-molecular","title":"MADGEN: Mass-Spec attends to De Novo Molecular generation","date":"2025-01-03","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":2,"samples_ran":2,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/jestr-joint-embedding-space-technique-for","title":"JESTR: Joint Embedding Space Technique for Ranking Candidate Molecules for the Annotation of Untargeted Metabolomics Data","date":"2024-11-18","rows_on_this_dataset":6,"code_links":0,"syntology":null},{"paper":"/paper/massspecgym-a-benchmark-for-the-discovery-and","title":"MassSpecGym: A benchmark for the discovery and identification of molecules","date":"2024-10-30","rows_on_this_dataset":24,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":6,"samples_ran":2,"samples_unverified":4,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":3,"samples_harvested":9,"samples_ran":5,"samples_unverified":4,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}