Hierarchical Clustering Analyses of Plasma Proteins in Subjects With Cardiovascular Risk Factors Identify Informative Subsets Based on Differential Levels of Angiogenic and Inflammatory Biomarkers

• Published in: Frontiers in neuroscience Vol. 14; p. 84
• Main Authors: Winder, Zachary, Sudduth, Tiffany L., Fardo, David, Cheng, Qiang, Goldstein, Larry B., Nelson, Peter T., Schmitt, Frederick A., Jicha, Gregory A., Wilcock, Donna M.
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 06.02.2020; Frontiers Media S.A
• Subjects: Angiogenesis; Biomarkers; Cardiovascular diseases; Cerebrovascular diseases; Cognitive ability; Datasets; Dementia; Fibroblast growth factors; Gelatinase B; hierarchical clustering analysis; IL8; Inflammation; Interleukin 8; Learning algorithms; Light; Machine learning; mild cognitive impairment; MMP1; Neuroscience; Plasma proteins; Risk factors; Studies; Tumor necrosis factor; vascular cognitive impairment and dementia; Vascular endothelial growth factor; VEGF; mild cognitive impairment; IL8; hierarchical clustering analysis; MMP1; VEGF; vascular cognitive impairment and dementia
• ISSN: 1662-453X; 1662-4548; 1662-453X
• DOI: 10.3389/fnins.2020.00084

Agglomerative hierarchical clustering analysis (HCA) is a commonly used unsupervised machine learning approach for identifying informative natural clusters of observations. HCA is performed by calculating a pairwise dissimilarity matrix and then clustering similar observations until all observations are grouped within a cluster. Verifying the empirical clusters produced by HCA is complex and not well studied in biomedical applications. Here, we demonstrate the comparability of a novel HCA technique with one that was used in previous biomedical applications while applying both techniques to plasma angiogenic (FGF, FLT, PIGF, Tie-2, VEGF, VEGF-D) and inflammatory (MMP1, MMP3, MMP9, IL8, TNFα) protein data to identify informative subsets of individuals. Study subjects were diagnosed with mild cognitive impairment due to cerebrovascular disease (MCI-CVD). Through comparison of the two HCA techniques, we were able to identify subsets of individuals, based on differences in VEGF ( < 0.001), MMP1 ( < 0.001), and IL8 ( < 0.001) levels. These profiles provide novel insights into angiogenic and inflammatory pathologies that may contribute to VCID.