Searching molecular structure databases with tandem mass spectra using CSI:FingerID

By Kai Dührkop, Huibin Shen, Marvin Meusel, Juho Rousu, Sebastian Böcker. Proceedings of the National Academy of Sciences of the United States of America, 2015

1,020 citations recorded by the source corpus.

Significance Untargeted metabolomics experiments usually rely on tandem MS (MS/MS) to identify the thousands of compounds in a biological sample. Today, the vast majority of metabolites remain unknown. Recently, several computational approaches were presented for searching molecular structure databases using MS/MS data. Here, we present CSI:FingerID, which combines fragmentation tree computation and machine learning. An in-depth evaluation on two large-scale datasets shows that our method can find 150% more correct identifications than the second-best search method. In comparison with the two runner-up methods, CSI:FingerID reaches 5.4-fold more unique identifications. We also present evaluations indicating that the performance of our method will further improve when more training data become available. CSI:FingerID is publicly available at www.csi-fingerid.org. Metabolites provide a direct functional signature of cellular state. Untargeted metabolomics experiments usually rely on tandem MS to identify the thousands of compounds in a biological sample. Today, the vast majority of metabolites remain unknown. We present a method for searching molecular structure databases using tandem MS data of small molecules. Our method computes a fragmentation tree that best explains the fragmentation spectrum of an unknown molecule. We use the fragmentation tree to predict the molecular structure fingerprint of the unknown compound using machine learning. This fingerprint is then used to search a molecular structure database such as PubChem. Our method is shown to improve on the competing methods for computational metabolite identification by a considerable margin.

Subfields: Computational Theory and Mathematics, Molecular Biology, Spectroscopy

View Paper: https://www.semanticscholar.org/paper/013c142c242b3239f5f19e41381b2175c7f3c1cd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

In Spectroscopy

Citations in the corpus

The Chemistry Development Kit (CDK) v2.0: atom typing, depiction, molecular formulas, and substructure searching (2017, background): “With the rise of metabolomics it has become increasingly relevant to have full support for these in cheminformatics libraries [23, 53–56].”
Software Tools and Approaches for Compound Identification of LC-MS/MS Data in Metabolomics (2018, methodology): “CSI:FingerID [98] is a freely available web-service and uses a two-step scheme: ﬁrst, a kernel-based approach is utilized to predict molecular ﬁngerprints [99] from its MS/MS spectrum and then the predicted molecular ﬁngerprints are matched against a molecular compound database.”
MolNetEnhancer: Enhanced Molecular Networks by Integrating Metabolome Mining and Annotation Tools (2019, methodology): “Complementary to library matching, in silico tools such as Network Annotation Propagation (NAP) [8], DEREPLICATOR [1], VarQuest [2], or SIRIUS+CSI:FingerID [4] predict fragmentation spectra in silico from known structures and allow for effective searching in chemical databases for candidate structur”
American Gut: an Open Platform for Citizen Science Microbiome Research (2018, positive): “In addition, commendamide (330.2640) and its analog ( m/z 344.2799) were identiﬁed by matching the exact mass of the corresponding ion and by in silico prediction of the MS/MS fragmentation spectra with CSI:FingerID (81) (Fig.”
Artificial Intelligence For Natural Product Drug Discovery (2023, methodology): “Since as early as 1960, AI has been used to complement rule-based approaches in de novo identification of unknown compounds from MS data 41,42 .”
MetFrag relaunched: incorporating strategies beyond in silico fragmentation (2016, methodology): “[18] investigated each individual fragmenter currently available and compared the results with Table 2 PubChem and ChemSpider results (number of pessimistic top 1 ranks) for 473 Eawag Orbitrap spectra”