A Multimodal Omics Exploration of the Motor and Non-Motor Symptoms of Parkinson’s Disease

Abstract

Parkinson’s disease (PD) is the second most common neurodegenerative disease clinically characterized by classical motor symptoms and a range of associated non-motor symptoms. Due to the heterogeneity of symptoms and variability in patient prognosis, the discovery of blood biomarkers is of utmost importance to identify the biological mechanisms underlying the different clinical manifestations of PD, monitor its progression and develop personalized treatment strategies. Whereas studies often rely on motor symptoms alone or composite scores, our study focused on finding relevant molecular markers associated with three clinical models describing either motor, cognitive or emotional symptoms. An integrative multiblock approach was performed using regularized generalized canonical correlation analysis to determine specific associations between lipidomics, transcriptomics and clinical data in 48 PD patients. We identified omics signatures confirming that clinical manifestations of PD in our cohort could be classified according to motor, cognition or emotion models. We found that immune-related genes and triglycerides were well-correlated with motor variables, while cognitive variables were linked to triglycerides as well as genes involved in neuronal growth, synaptic plasticity and mitochondrial fatty acid oxidation. Furthermore, emotion variables were associated with phosphatidylcholines, cholesteryl esters and genes related to endoplasmic reticulum stress and cell regulation.

Automatic pretreatment and multiblock analysis of flavor release and sensory temporal data simultaneously collected in vivo

“…The regularized generalized canonical correlation analysis (RGCCA) is a multiblock method that was introduced by Tenenhaus and Tenenhaus 13 as a general framework for a large panel of multiblock methods. It was mainly used for the research of multimodal biomarkers as presented in Xicota et al 14 or Lejeune et al 15 …”