Fig. 3: Idiopathic PD endotypes.
From: A proteogenomic view of Parkinson’s disease causality and heterogeneity
A Heatmap of z-scores of the protein values as measured with SomaScan, corresponding to the two modules identified using Weighted Gene Co-expression Network Analysis (WGCNA). The proteins in these modules are used for cluster analysis using Consensus Clustering, which retrieves two clusters (endotypes) of idiopathic PD patients. Patients are shown in the x-axis, separated by endotype, while proteins are shown in the y-axis, separated by module. B Tracking plot depicting how the idiopathic PD patients are assigned to specific endotypes by Consensus Clustering as the number of potential endotypes increases. C Partition tree predicting endotype membership of the idiopathic PD patients based on clinical variables only. One node suffices to separate patients into endotypes based on phospho-tau levels (p-tau) in CSF as measured with a clinical assay, the cut-off being 11 pg/mL.
