This paper presents maplet, an open-source R package for the creation of highly customizable, fully reproducible statistical pipelines for metabolomics data analysis. It builds on the SummarizedExperiment data structure to create a centralized pipeline framework for storing data, analysis steps, results, and visualizations. maplet’s key design feature is its modularity, which offers several advantages, such as ensuring code quality through the maintenance of individual functions and promoting collaborative development by removing technical barriers to code contribution. With over 90 functions, the package includes a wide range of functionalities, covering many widely used statistical approaches and data visualization techniques. Availability The maplet package is implemented in R and freely available at https://github.com/krumsieklab/maplet.

Dimensionality reduction approaches are commonly used for the deconvolution of high-dimensional metabolomics datasets into underlying core metabolic processes. However, current state-of-the-art methods are widely incapable of detecting nonlinearities in metabolomics data. Variational Autoencoders (VAEs) are a deep learning method designed to learn nonlinear latent representations which generalize to unseen data. Here, we trained a VAE on a large-scale metabolomics population cohort of human blood samples consisting of over 4500 individuals. We analyzed the pathway composition of the latent space using a global feature importance score, which demonstrated that latent dimensions represent distinct cellular processes. To demonstrate model generalizability, we generated latent representations of unseen metabolomics datasets on type 2 diabetes, acute myeloid leukemia, and schizophrenia and found significant correlations with clinical patient groups. Notably, the VAE representations showed stronger effects than latent dimensions derived by linear and non-linear principal component analysis. Taken together, we demonstrate that the VAE is a powerful method that learns biologically meaningful, nonlinear, and transferrable latent representations of metabolomics data.

Alzheimer's disease (AD) is a major neurodegenerative disorder with significant environmental factors, including diet and lifestyle, influencing its onset and progression. Although previous studies have suggested that certain diets may reduce the incidence of AD, the underlying mechanisms remain unclear. In this randomized crossover study of 20 elderly adults, we investigated the effects of a modified Mediterranean ketogenic diet (MMKD) on the plasma lipidome, analyzing 784 lipid species across 47 classes. We identified substantial changes in response to intervention, notably a global elevation across all plasmanyl and plasmenyl ether lipid species among others, with many changes linked with changes to clinical and biochemical markers of AD. Leveraging our prior clinical studies on AD (n = 1,912), we found that MMKD was inversely associated with the lipidomic signature of prevalent and incident AD. Given its low risk and cost, MMKD could be a promising approach for prevention or early symptomatic treatment of AD.

Dimensionality reduction approaches are commonly used for the deconvolution of high-dimensional metabolomics datasets into underlying core metabolic processes. However, current state-of-the-art methods are widely incapable of detecting nonlinearities in metabolomics data. Variational Autoencoders (VAEs) are a deep learning method designed to learn nonlinear latent representations which generalize to unseen data. Here, we trained a VAE on a large-scale metabolomics population cohort of human blood samples consisting of over 4,500 individuals. We analyzed the pathway composition of the latent space using a global feature importance score, which showed that latent dimensions represent distinct cellular processes. To demonstrate model generalizability, we generated latent representations of unseen metabolomics datasets on type 2 diabetes, schizophrenia, and acute myeloid leukemia and found significant correlations with clinical patient groups. Taken together, we demonstrate for the first time that the VAE is a powerful method that learns biologically meaningful, nonlinear, and universal latent representations of metabolomics data.