FlexMS is a flexible framework for benchmarking deep learning-based mass spectrum prediction tools in metabolomics

TL;DR

FlexMS is a versatile benchmarking framework designed to evaluate various deep learning models for mass spectrum prediction in metabolomics, facilitating better model comparison and selection.

Contribution

We developed FlexMS, an easy-to-use framework that enables dynamic construction and assessment of diverse deep learning models for mass spectrum prediction.

Findings

Performance varies with dataset diversity and hyperparameters.

Pretraining and transfer learning significantly impact accuracy.

Retrieval benchmarks help in practical molecule identification.

Abstract

The identification and property prediction of chemical molecules is of central importance in the advancement of drug discovery and material science, where the tandem mass spectrometry technology gives valuable fragmentation cues in the form of mass-to-charge ratio peaks. However, the lack of experimental spectra hinders the attachment of each molecular identification, and thus urges the establishment of prediction approaches for computational models. Deep learning models appear promising for predicting molecular structure spectra, but overall assessment remains challenging as a result of the heterogeneity in methods and the lack of well-defined benchmarks. To address this, our contribution is the creation of benchmark framework FlexMS for constructing and evaluating diverse model architectures in mass spectrum prediction. With its easy-to-use flexibility, FlexMS supports the dynamic construction of numerous distinct combinations of model architectures, while assessing their performance on preprocessed public datasets using different metrics. In this paper, we provide insights into factors influencing performance, including the structural diversity of datasets, hyperparameters like learning rate and data sparsity, pretraining effects, metadata ablation settings and cross-domain transfer learning analysis. This provides practical guidance in choosing suitable models. Moreover, retrieval benchmarks simulate practical identification scenarios and score potential matches based on predicted spectra.