Fig. 3

Pathway analysis of 76 selected differentially expressed genes with emphasis on three subnetworks. Using the online program STRING-db, we generated a broad interaction network from our selected 76 genes. Within the network, we assigned particular colors to each node (gene) to represent the expression of that particular gene in patients with SS. Node colors are explained in the figure. The yellow nodes are originally derived from previous genome-wide association studies (GWAS) of patients with Sjögren’s syndrome (SS) and are added for the validation and strengthening of our independently derived network. The red and purple nodes represent genes that are upregulated and downregulated in SS, respectively. Similarly, the light blue and green nodes represent genes exhibiting upregulation and downregulation in SS, respectively. These genes have not yet been strongly linked to SS pathophysiology. The lines connecting the nodes (edges) represent interactions between two nodes and can be derived from evidence or inferred from previously known data: magenta experimentally derived; blue predicted interaction through gene co-occurrence; gray predicted functional interaction derived from either homologous protein interactions in other species/associations in curated databases/co-mentioned in PubMed abstracts; yellow transcriptional regulation (experimentally derived); black reaction (experimentally derived); and purple catalysis (experimentally derived). Edges ending in a green arrow, red bar, or black circle represent an action between the two nodes that can range from positive, negative, or unspecified, respectively. Subnetworks: A genes of the interferon gene family and interferon-stimulated genes are outlined by a black dashed line; B genes associated with chemotaxis are outlined by a black dashed line; C genes identified in our study as potential biomarkers of SS are outlined by a light green dashed line