@article{HuberRidderVerhoevenetal.2021, author = {Huber, Florian and Ridder, Lars and Verhoeven, Stefan and Spaaks, Jurriaan H. and Diblen, Faruk and Rogers, Simon and van der Hooft, Justin J. J.}, title = {Spec2Vec: Improved mass spectral similarity scoring through learning of structural relationships}, series = {PLOS Computational Biology}, volume = {17}, journal = {PLOS Computational Biology}, number = {2}, publisher = {Cold Spring Harbor Laboratory}, organization = {PLOS}, issn = {1553-7358}, doi = {10.1371/journal.pcbi.1008724}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:due62-opus-34687}, year = {2021}, abstract = {Spectral similarity is used as a proxy for structural similarity in many tandem mass spectrometry (MS/MS) based metabolomics analyses such as library matching and molecular networking. Although weaknesses in the relationship between spectral similarity scores and the true structural similarities have been described, little development of alternative scores has been undertaken. Here, we introduce Spec2Vec, a novel spectral similarity score inspired by a natural language processing algorithm-Word2Vec. Spec2Vec learns fragmental relationships within a large set of spectral data to derive abstract spectral embeddings that can be used to assess spectral similarities. Using data derived from GNPS MS/MS libraries including spectra for nearly 13,000 unique molecules, we show how Spec2Vec scores correlate better with structural similarity than cosine-based scores. We demonstrate the advantages of Spec2Vec in library matching and molecular networking. Spec2Vec is computationally more scalable allowing structural analogue searches in large databases within seconds.}, language = {en} } @article{SchornVerhoevenRidderetal.2021, author = {Schorn, Michelle A. and Verhoeven, Stefan and Ridder, Lars and Huber, Florian and Acharya, Deepa D. and Aksenov, Alexander A. and Aleti, Gajender and Moghaddam, Jamshid Amiri and Aron, Allegra T. and Aziz, Saefuddin and Bauermeister, Anelize and Bauman, Katherine D. and Baunach, Martin and Beemelmanns, Christine and Beman, J. Michael and Berlanga-Clavero, Mar{\´i}a Victoria and Blacutt, Alex A. and Bode, Helge B. and Boullie, Anne and Brejnrod, Asker and Bugni, Tim S. and Calteau, Alexandra and Cao, Liu and Carri{\´o}n, V{\´i}ctor J. and Castelo-Branco, Raquel and Chanana, Shaurya and Chase, Alexander B. and Chevrette, Marc G. and Costa-Lotufo, Leticia V. and Crawford, Jason M. and Currie, Cameron R. and Cuypers, Bart and Dang, Tam and de Rond, Tristan and Demko, Alyssa M. and Dittmann, Elke and Du, Chao and Drozd, Christopher and Dujardin, Jean-Claude and Dutton, Rachel J. and Edlund, Anna and Fewer, David P. and Garg, Neha and Gauglitz, Julia M. and Gentry, Emily C. and Gerwick, Lena and Glukhov, Evgenia and Gross, Harald and Gugger, Muriel and Guill{\´e}n Matus, Dulce G. and Helfrich, Eric J. N. and Hempel, Benjamin-Florian and Hur, Jae-Seoun and Iorio, Marianna and Jensen, Paul R. and Kang, Kyo Bin and Kaysser, Leonard and Kelleher, Neil L. and Kim, Chung Sub and Kim, Ki Hyun and Koester, Irina and K{\"o}nig, Gabriele M. and Leao, Tiago and Lee, Seoung Rak and Lee, Yi-Yuan and Li, Xuanji and Little, Jessica C. and Maloney, Katherine N. and M{\"a}nnle, Daniel and Martin H, Christian and McAvoy, Andrew C. and Metcalf, Willam W. and Mohimani, Hosein and Molina-Santiago, Carlos and Moore, Bradley S. and Mullowney, Michael W. and Muskat, Mitchell and Nothias, Louis-F{\´e}lix and O'Neill, Ellis C. and Parkinson, Elizabeth I. and Petras, Daniel and Piel, J{\"o}rn and Pierce, Emily C. and Pires, Karine and Reher, Raphael and Romero, Diego and Roper, M. Caroline and Rust, Michael and Saad, Hamada and Saenz, Carmen and Sanchez, Laura M. and S{\o}rensen, S{\o}ren Johannes and Sosio, Margherita and S{\"u}ssmuth, Roderich D. and Sweeney, Douglas and Tahlan, Kapil and Thomson, Regan J. and Tobias, Nicholas J. and Trindade-Silva, Amaro E. and van Wezel, Gilles P. and Wang, Mingxun and Weldon, Kelly C. and Zhang, Fan and Ziemert, Nadine and Duncan, Katherine R. and Cr{\"u}semann, Max and Rogers, Simon and Dorrestein, Pieter C. and Medema, Marnix H. and van der Hooft, Justin J. J.}, title = {A community resource for paired genomic and metabolomic data mining}, series = {Nature Chemical Biology}, volume = {17}, journal = {Nature Chemical Biology}, number = {4}, publisher = {Nature}, issn = {1552-4469}, doi = {10.1038/s41589-020-00724-z}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:due62-opus-34708}, pages = {363 -- 368}, year = {2021}, language = {en} } @article{BeniddirKangGentaJouveetal.2021, author = {Beniddir, Mehdi A. and Kang, Kyo Bin and Genta-Jouve, Gr{\´e}gory and Huber, Florian and Rogers, Simon and van der Hooft, Justin J. J.}, title = {Advances in decomposing complex metabolite mixtures using substructure- and network-based computational metabolomics approaches}, series = {Natural Product Reports}, volume = {38}, journal = {Natural Product Reports}, number = {11}, publisher = {The Royal Society of Chemistry}, issn = {1460-4752}, doi = {10.1039/D1NP00023C}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:due62-opus-34772}, pages = {1967 -- 1993}, year = {2021}, language = {en} } @article{HubervanderBurgvanderHooftetal.2021, author = {Huber, Florian and van der Burg, Sven and van der Hooft, Justin J. J. and Ridder, Lars}, title = {MS2DeepScore: a novel deep learning similarity measure to compare tandem mass spectra}, series = {Journal of Cheminformatics}, volume = {13}, journal = {Journal of Cheminformatics}, number = {1}, publisher = {Cold Spring Harbor Laboratory}, issn = {1758-2946}, doi = {10.1186/s13321-021-00558-4}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:due62-opus-34847}, pages = {84}, year = {2021}, abstract = {Mass spectrometry data is one of the key sources of information in many workflows in medicine and across the life sciences. Mass fragmentation spectra are generally considered to be characteristic signatures of the chemical compound they originate from, yet the chemical structure itself usually cannot be easily deduced from the spectrum. Often, spectral similarity measures are used as a proxy for structural similarity but this approach is strongly limited by a generally poor correlation between both metrics. Here, we propose MS2DeepScore: a novel Siamese neural network to predict the structural similarity between two chemical structures solely based on their MS/MS fragmentation spectra. Using a cleaned dataset of > 100,000 mass spectra of about 15,000 unique known compounds, we trained MS2DeepScore to predict structural similarity scores for spectrum pairs with high accuracy. In addition, sampling different model varieties through Monte-Carlo Dropout is used to further improve the predictions and assess the model's prediction uncertainty. On 3600 spectra of 500 unseen compounds, MS2DeepScore is able to identify highly-reliable structural matches and to predict Tanimoto scores for pairs of molecules based on their fragment spectra with a root mean squared error of about 0.15. Furthermore, the prediction uncertainty estimate can be used to select a subset of predictions with a root mean squared error of about 0.1. Furthermore, we demonstrate that MS2DeepScore outperforms classical spectral similarity measures in retrieving chemically related compound pairs from large mass spectral datasets, thereby illustrating its potential for spectral library matching. Finally, MS2DeepScore can also be used to create chemically meaningful mass spectral embeddings that could be used to cluster large numbers of spectra. Added to the recently introduced unsupervised Spec2Vec metric, we believe that machine learning-supported mass spectral similarity measures have great potential for a range of metabolomics data processing pipelines.}, language = {en} }