Related Experiment Video
Updated: Apr 28, 2026
11:03
Massively Parallel Reporter Assays in Cultured Mammalian Cells
Published on: August 17, 2014
22.3K
Inherent variability limits clinical utility of reproducible Parkinson's transcriptomics signatures
Roy Dayan1,2, Serafima Dubnov3, Hagit Turm3,4
1Department of Neurology, Hadassah Medical Center, Jerusalem, Israel. ROYD@hadassah.org.il.
NPJ Parkinson'S Disease
|December 18, 2025
Summary
Blood gene expression signatures show limited clinical utility for diagnosing Parkinson's disease (PD). Despite testing 13 signatures, classification performance remained low, suggesting challenges in developing a universal diagnostic tool.
Area of Science:
- Genomics and Molecular Biology
- Neurodegenerative Diseases
- Biomarker Discovery
Background:
- Blood transcriptomic signatures have been explored for Parkinson's disease (PD) diagnosis for decades.
- Previous translational efforts have not led to clinical integration of these RNA-based diagnostic markers.
- Coding RNA signatures are a focus for potential PD diagnostic biomarkers.
Purpose of the Study:
- To evaluate the diagnostic classification performance of 13 published coding RNA-based signatures for Parkinson's disease.
- To assess signature performance on both a large public dataset and a prospectively collected clinical dataset.
- To investigate factors influencing signature performance, including gene overlap and GBA1 association.
Main Methods:
- Utilized 13 published coding RNA-based transcriptomic signatures.
- Tested signatures on the Parkinson's Progression Markers Initiative (PPMI) public dataset.
- Collected and analyzed prospective data from levodopa-naïve PD patients and healthy controls under controlled conditions.
Main Results:
- Low but significant gene overlap (2.7%) was observed between signatures, enriched for lipid metabolism genes.
- Most signatures (10/13) showed significance on the PPMI dataset, but with lower performance (median AUC: 59.7%) than previously reported.
- Classification performance improved for GBA1-associated PD, but rigorous standardization in a prospective study did not enhance overall transcriptome-based classification.
Conclusions:
- The search for a universal blood-based transcriptomic signature for Parkinson's disease diagnosis faces significant limitations in clinical utility.
- While these signatures may offer insights into PD pathophysiology, their current performance does not support widespread clinical application.
- Future research may need to focus on more specific or combined biomarker approaches rather than universal transcriptomic signatures.
