MolDiscovery: Learning Mass Spectrometry Fragmentation of Small Molecules

Abstract

Identification of small molecules is a critical task in various areas of life science. Recent advances in mass spectrometry have enabled the collection of tandem mass spectra of small molecules from hundreds of thousands of environments. To identify which molecules are present in a sample, one can search mass spectra collected from the sample against millions of molecular structures in small molecule databases. This is a challenging task as currently it is not clear how small molecules are fragmented in mass spectrometry. The existing approaches use the domain knowledge from chemistry to predict fragmentation of molecules. However, these rule-based methods fail to explain many of the peaks in mass spectra of small molecules. Recently, spectral libraries with tens of thousands of labelled mass spectra of small molecules have emerged, paving the path for learning more accurate fragmentation models for mass spectral database search. We present molDiscovery, a mass spectral database search method that improves both efficiency and accuracy of small molecule identification by (i) utilizing an efficient algorithm to generate mass spectrometry fragmentations, and (ii) learning a probabilistic model to match small molecules with their mass spectra. We show our database search is an order of magnitude more efficient than the state-of-the-art methods, which enables searching against databases with millions of molecules. A search of over 8 million spectra from the Global Natural Product Social molecular networking infrastructure shows that our probabilistic model can correctly identify nearly six times more unique small molecules than previous methods. Moreover, by applying molDiscovery on microbial datasets with both mass spectral and genomics data we successfully discovered the novel biosynthetic gene clusters of three families of small molecules.AvailabilityThe command-line version of molDiscovery and its online web service through the GNPS infrastructure are available at https://github.com/mohimanilab/molDiscovery.

Assigning confidence to structural annotations from mass spectra with COSMIC

“…So-called "in silico methods" allow us to search in a molecular structure database using MS/MS data as our query. Most methods follow one of three paradigms: (i) Combinatorial fragmenters 13,[47][48][49] try to explain the query spectrum using the candidate structure, combinatorially breaking bonds in the molecular structure graph. (ii) Other methods try to predict the fragmentation spectrum of a given compound structure 14,50,51 ; this allows us to search in the structure database by spectral matching.…”

Metabolome Exploration by High-Resolution Mass Spectrometry Methodologies of Two New Yeast Species: Starmerella reginensis and Starmerella kourouensis

“…After experimenting several bioinformatic tools on metabolomic datasets, we chose to apply the Network Annotation Propagation (NAP) and the MolDiscovery tool for their ease of use and for the most exhaustive and rapid exploration possible. , …”

“…22,23 Additional tools, such as Network Annotation Propagation (NAP) or MolDiscovery, use all the capacities of spectral databases and in silico fragmentation principles to accelerate the dereplication process and facilitate the manual investigation of acquired data. 24,25 All the annotation provided by this treatment may be gauged by the annotation level system of the Metabolomics Standard Initiative. 26 This system uses four levels covering fully identified molecules, putatively annotated compounds, putatively annotated compound chemical classes, and unknown metabolites.…”

“…The accurate mass measurements of detected features are compared with spectral databases (e.g., Human Metabolome Database and MassBank , ) to annotate the compounds. MZMine2-preprocessed MS/MS data may be used to build molecular networks on the Global Natural Product Social (GNPS) molecular networking platform () as structurally related molecules share similar fragmentation patterns. , Additional tools, such as Network Annotation Propagation (NAP) or MolDiscovery, use all the capacities of spectral databases and in silico fragmentation principles to accelerate the dereplication process and facilitate the manual investigation of acquired data. , All the annotation provided by this treatment may be gauged by the annotation level system of the Metabolomics Standard Initiative . This system uses four levels covering fully identified molecules, putatively annotated compounds, putatively annotated compound chemical classes, and unknown metabolites.…”

Metabolomics of Healthy and Stony Coral Tissue Loss Disease Affected Montastraea cavernosa Corals

“…Additional compound annotations were performed using the XCMS Online, and in silico tools such as MolDiscovery (v.1.0.0) and SIRIUS 4.5 with CSI:FingerID and CANOPUS, and through literature searches for compounds known to be produced by corals and their endosymbionts for untargeted analysis. MolDiscovery, available through the GNPS platform, compares in silico generated mass spectra of small molecules from a variety of databases with experimental MS 2 spectra and includes a similarity score for each reported match (Cao et al, 2020). The LC-MS feature list from the extracts of coral samples collected in October (mgf file) was submitted to the MolDiscovery workflow.…”