AIPD-DC8-RCSI

• Host Institution: RCSI University of Medicine & Health Sciences
• PhD Enrolment: RCSI University of Medicine & Health Sciences
  
• Start Date: October 2025
  
• Duration: 36 months
• Official PhD Supervisor: Jochen Prehn

Research Objectives

Mitochondrial dysfunction and alterations in energy and lipid metabolism are hallmarks of PD. The aim of this project is to perform a metabolic analysis of blood-based expression profiles to derive clinically accessible, peripheral biomarker signatures based on metabolic dysfunction for personalised PD disease monitoring and targeting.

Specific objectives:

1. Identify metabolic signatures in the blood of PD patients and their relation to diagnosis, symptoms and disease progression;
2. Predict altered metabolic fluxes/pathways in individual patients by incorporation of gene expression profiles and network-wide analysis;
3. Identify metabolic reactions/nodes as putative therapeutic targets;
4. Incorporate clinical markers to derive enhanced signatures of metabolic dysfunction for disease monitoring.

The student will first determine differentially expressed metabolites and will perform correlation analyses on metabolomics datasets derived from peripheral blood mononuclear cells (PBMCs) from LuxPARK to describe the heterogeneity of metabolic signatures in PD patients. Signatures/biomarkers will be correlated with diagnosis (IPD, PDD, DLB etc), symptoms (motor, non-motor) and disease stage. Next, at RCSI, the student will be trained in the application of systems biology tools such as genome-scale metabolic models (GSMM) and will generate GSMMs of PBMC/blood gene and/or protein expression profiles to derive functional insight into the distinct metabolic signatures, and predict individual, specifically altered metabolic fluxes/pathways and their relation to diagnosis, symptoms and disease progression. Blood- and brain-based models will be aligned to determine shared and distinct patterns of metabolic dysfunction. Both publicly available datasets as well as brain transcriptomic, proteomic, and metabolomic data generated by the IMI2 PDMitoQUANT project will be available. Knockin/ knock-out and perturbation analyses in silico will predict critical metabolic reactions/nodes and putative therapeutic targets. At UL student will apply AI/ML and XAI techniques to combine clinical features as well as genomic mutations to derive enhanced signatures of metabolic dysfunction to further dissect specific pathways for therapeutic monitoring. Signatures/biomarkers will be validated using PPMI. Together, derived metabolic biomarkers will enable a refined diagnosis of PD patients and allow for longitudinal, personalised monitoring of symptoms. DC8 will closely work with DC14 due to the common focus on metabolic signatures.

Expected Results

• AI/ML models and biomarker signatures for metabolic disease monitoring/staging
• Putative therapeutic targets
   

Planned Secondment(s)

• Host 1: Centre Hospitalier de Luxembourg
  
  • Duration: 18 months
    
  • Purpose: Learning about PD from a clinical perspective
• Host 2: University of Luxembourg
  
  • Duration: 9 months
    
  • Purpose: Learning about the development of AI/ML models

This project is part of the "Precision Neurology" work package.