Analysis of high-dimensional metabolomics data with complex temporal dynamics using RM-ASCA+

The intricate dependency structure of biological “omics” data, particularly those originating from longitudinal intervention studies with frequently sampled repeated measurements renders the analysis of such data challenging. The high-dimensionality, inter-relatedness of multiple outcomes, and heterogeneity in the studied systems all add to the difficulty in deriving meaningful information. In addition, the subtle differences in dynamics often deemed meaningful in nutritional intervention studies can be particularly challenging to quantify. In this work we demonstrate the use of quantitative longitudinal models within the repeated-measures ANOVA simultaneous component analysis+ (RM-ASCA+) framework to capture the dynamics in frequently sampled longitudinal data with multivariate outcomes. We illustrate the use of linear mixed models with polynomial and spline basis expansion of the time variable within RM-ASCA+ in order to quantify non-linear dynamics in a simulation study as well as in a metabolomics data set. We show that the proposed approach presents a convenient and interpretable way to systematically quantify and summarize multivariate outcomes in longitudinal studies while accounting for proper within subject dependency structures.Author summary: With advances in high-throughput omics platforms coupled with a reduction in associated costs, we increasingly see intervention studies generating extensive time-series of measurements simultaneously capturing changes across many dimensions. However, in order to derive meaningful information from such data, we must take into account the high-dimensionality, the interrelatedness of outcomes, the experimental design, the temporal dependency, as well as the subject-to-subject variability. Analytical tools that are able to account for all of these properties, in particular approaches that incorporate the temporal dependencies and the corresponding between-subject variability, are needed to make efficient use of such data. Here, we introduce novel methodology to quantify the temporal dependency and its subject-to-subject variability in high-dimensional, frequently sampled time-series data from longitudinal intervention studies. Our approach provides a systematic way to quantify and summarize multivariate outcomes in longitudinal intervention studies while accounting for the study design, the temporal dependency, and its between-subject variability. We demonstrate the effectiveness of this approach in a simulation study as well as on a metabolomics dataset.