A dynamic probabilistic principal components model for the analysis of longitudinal metabolomic data

TL;DR

This paper introduces a dynamic probabilistic PCA model tailored for longitudinal metabolomic data, effectively reducing dimensionality and identifying influential metabolites over time by modeling correlations with autoregressive structures.

Contribution

The novel DPPCA model combines probabilistic PCA with autoregressive dynamics to handle longitudinal data, enhancing dimension reduction and metabolite influence detection.

Findings

Successfully applied to animal metabolomics data

Effectively captures temporal correlations in metabolites

Identifies key metabolites changing over time

Abstract

In a longitudinal metabolomics study, multiple metabolites are measured from several observations at many time points. Interest lies in reducing the dimensionality of such data and in highlighting influential metabolites which change over time. A dynamic probabilistic principal components analysis (DPPCA) model is proposed to achieve dimension reduction while appropriately modelling the correlation due to repeated measurements. This is achieved by assuming an autoregressive model for some of the model parameters. Linear mixed models are subsequently used to identify influential metabolites which change over time. The proposed model is used to analyse data from a longitudinal metabolomics animal study.