Convolutional Neural Network-Based Compound Fingerprint Prediction for Metabolite Annotation

Abstract

Metabolite annotation has been a challenging issue especially in untargeted metabolomics studies by liquid chromatography coupled with mass spectrometry (LC-MS). This is in part due to the limitations of publicly available spectral libraries, which consist of tandem mass spectrometry (MS/MS) data acquired from just a fraction of known metabolites. Machine learning provides the opportunity to predict molecular fingerprints based on MS/MS data. The predicted molecular fingerprints can then be used to help rank putative metabolite IDs obtained by using either the precursor mass or the formula of the unknown metabolite. This method is particularly useful to help annotate metabolites whose corresponding MS/MS spectra are missing or cannot be matched with those in accessible spectral libraries. We investigated a convolutional neural network (CNN) for molecular fingerprint prediction based on data acquired by MS/MS. We used more than 680,000 MS/MS spectra obtained from the MoNA repository and NIST 20, representing about 36,000 compounds for training and testing our CNN model. The trained CNN model is implemented as a python package, MetFID. The package is available on GitHub for users to enter their MS/MS spectra and corresponding putative metabolite IDs to obtain ranked lists of metabolites. Better performance is achieved by MetFID in ranking putative metabolite IDs using the CASMI 2016 benchmark dataset compared to two other machine learning-based tools (CSI:FingerID and ChemDistiller).

Machine learning-assisted structure annotation of natural products based on MS and NMR data

“…Compared to CSI:FingerID, the improved MetFID achieves better performance in ranking putative metabolite IDs on the CASMI 2016 benchmark dataset. 77 Moreover, there are also comparable algorithms, such as DeepEI which utilizes a DNN framework for the prediction of molecular fingerprint. 64 Recently, Lang et al tried to omit the step of converting MS image into molecular fingerprint, and directly analyzed the pixel image of MS/MS images with CNN.…”

Recent advances in data- and knowledge-driven approaches to explore primary microbial metabolism

“…Gao et al. took the approach of predicting molecular fingerprints directly from spectral information using convolutional NNs trained on spectra of more than 36,000 compounds from public databases [ 42 ]. MSNovelist predicts de novo compound structures from tandem mass spectra with recurrent NNs [ 43 ], while Retip offers a set of machine learning models to predict retention time of compounds, given an experimental training set collected by the user [ 44 ] ( Figure 1 ).…”