Probabilistic metabolite annotation using retention time prediction and meta-learned projections

Abstract

Retention time information is used for metabolite annotation in metabolomic experiments. But its usefulness is hindered by the availability of experimental retention time data in metabolomic databases, and by the lack of reproducibility between different chromatographic methods. Accurate prediction of retention time for a given chromatographic method would be a valuable support for metabolite annotation. We have trained state-of-the-art machine learning regressors using the 80, 038 experimental retention times from the METLIN Small Molecule Retention Tim (SMRT) dataset. The models included deep neural networks, deep kernel learning, several gradient boosting models, and a blending approach. 5, 666 molecular descriptors and 2, 214 fingerprints (MACCS166, Extended Connectivity, and Path Fingerprints fingerprints) were generated with the alvaDesc software. The models were trained using only the descriptors, only the fingerprints, and both types of features simultaneously. Bayesian hyperparameter search was used for parameter tuning. To avoid data-leakage when reporting the performance metrics, nested cross-validation was employed. The best results were obtained by a heavily regularized deep neural network trained with cosine annealing warm restarts and stochastic weight averaging, achieving a mean and median absolute errors of $$39.2 \pm 1.2\; s$$ 39.2 ± 1.2 s and $$17.2 \pm 0.9\;s$$ 17.2 ± 0.9 s , respectively. To the best of our knowledge, these are the most accurate predictions published up to date over the SMRT dataset. To project retention times between chromatographic methods, a novel Bayesian meta-learning approach that can learn from just a few molecules is proposed. By applying this projection between the deep neural network retention time predictions and a given chromatographic method, our approach can be integrated into a metabolite annotation workflow to obtain z-scores for the candidate annotations. To this end, it is enough that just as few as 10 molecules of a given experiment have been identified (probably by using pure metabolite standards). The use of z-scores permits considering the uncertainty in the projection when ranking candidates, and not only the accuracy. In this scenario, our results show that in 68% of the cases the correct molecule was among the top three candidates filtered by mass and ranked according to z-scores. This shows the usefulness of this information to support metabolite annotation. Python code is available on GitHub at https://github.com/constantino-garcia/cmmrt.

Good practices and recommendations for using and benchmarking computational metabolomics metabolite annotation tools

“…These strategies have in common that they mainly rely on MS/MS data, i.e., fragment peaks and intensities. However, recently, additional and often complementary information such as instrument type or collision energy, retention time or order are also utilised for ML model training (Bach et al, 2022 ; García et al, 2022 ; Witting & Böcker, 2020 ).

Fig.

…”

“…These strategies have in common that they mainly rely on MS/MS data, i.e., fragment peaks and intensities. However, recently, additional and often complementary information such as instrument type or collision energy, retention time or order are also utilised for ML model training (Bach et al, 2022;García et al, 2022;Witting & Böcker, 2020). The first strategy is generally not aimed at immediate metabolite annotation, but rather to translate MS/MS spectra into abstract representations that still are chemically meaningful although likely not understandable for anything but the trained model.…”

“…Findable, Accessible, Interoperable, and Reusable) data sharing initiatives (Neumann, 2022), researchers and tool developers are starting to have uncomplicated and straight-forward access to structural and mass spectral data of myriads of molecules. This has fuelled the large-scale and repository-scale reuse of mass spectrometry data (Haug et al, 2019;Jarmusch et al, 2020Jarmusch et al, , 2021Sud et al, 2016), now also including retention time data (García et al, 2022). Furthermore, the adoption of standards such as the Universal Spectrum Identifier (USI) (Deutsch et al, 2021) will further aid in the harmonization of efforts, making access to large amounts of data for training and validation purposes much more straightforward.…”

Contribution of allergy in the acquisition of uncontrolled severe asthma

“…Metabolite annotation was performed by comparison with online platform CEU Mass Mediator 3.0 ( 29 – 32 ) for data bases, and confirmed through tandem mass spectrometry (MS/MS). LC-MS/MS was performed in a similar LC-Q-ToF-MS instrument than the original experiment (Agilent series 1290 HPLC and series 6550 Q-ToF, respectively); and the method for the new equipment, which has been previously described ( 33 , 34 ), was adapted to be as close as possible to the original method used.…”

Generic and accurate prediction of retention times in liquid chromatography by post-projection calibration

“…To the best of our knowledge, the largest experimental RT dataset at present is METLIN (http://metlin.scripps.edu) containing 80,038 molecules, which was determined on the Zorbax Extend-C18 reverse-phase (RP) column by using 0.1% formic acid in water and acetonitrile as mobile phases. This dataset has been reported multiple times for establishing structure-based models, with a median prediction error of 1.6% −6.2% [19][20][21][22] . However, these models usually worked for speci c LC setups, mainly due to the variance in RT between different CMs.…”