University of Maryland School of Medicine Baltimore, Maryland, United States
Objectives: Despite growing interest in blood-based biomarkers for stroke recovery, there is no clear consensus on which molecular profiles should be captured. This study aims to identify key protein-protein interactions and biological pathways associated with stroke recovery using network analysis with a focus on biomarkers accessible in peripheral blood.
Design: Protein associations with the terms “stroke” and “recovery” were retrieved from STRING v12.0 for Homo sapiens using the embedded PubMed query function and a minimum confidence score of 0.7. Interaction sources included text mining, experimental evidence, curated databases, and co-expression. The resulting network was imported into Cytoscape v3.10.2 for topological and enrichment analysis. Network metrics were calculated to identify hub proteins. Functional enrichment was performed within Cytoscape to highlight Gene Ontology (GO) biological processes. Nodes were plotted on a log-scaled curve based on betweenness centrality to visualize network structure and identify hub nodes.
Results: The query yielded 97 protein associations. Influential nodes—defined by high betweenness centrality above the inflection point—included TNF, BDNF, GFAP, PTPRC, MBP, NTRK2, ITGAM, IL18, MTOR, and APOE.The query yielded 97 protein associations with “stroke” and “recovery.” Influential nodes—above the inflection point in the log curve where centrality scores plateaued—included TNF, BDNF, GFAP, PTPRC, MBP, NTRK2, ITGAM, IL18, MTOR, and APOE. Functional enrichment analysis identified several key biological processes with significant FDR-adjusted p-values. These included positive regulation of molecular function (FDR = 1.33E–5), positive regulation of neurogenesis (FDR = 7.54E–5), regulation of neuron projection development (FDR = 4.9E–4), positive regulation of leukocyte-mediated immunity (FDR = 2.9E–4), regulation of amyloid-beta formation (FDR = 4.1E–4), and BDNF receptor signaling pathway (FDR = 0.001).
Conclusions: This study identifies proteins accessible in peripheral blood that are central to stroke recovery–associated networks. Network analysis offers a reproducible approach to identifying candidate biomarkers that can support future efforts to integrate blood-based profiling into precision rehabilitation strategies.
