Fig. 3: Cluster enrichment analysis result of the top 500-proximal gene network.

a–c Significantly enriched terms (B-H q-value ≤ 0.05) identified in cluster enrichment analysis of the top 500-proximal gene network derived from the 41 protein-coding DE genes. The cluster analysis identified three clusters: 0 (blue), 1 (orange), and 2 (green), containing 239, 168, and 93 genes, respectively. A total 189 functional terms were significantly enriched (B-H q-value ≤ 0.05) among the three clusters. Overall, cluster 0 had the most enriched terms, of these, the term with the greatest enrichment was ‘extracellular matrix’ (GO:0031012; B-H q = 1.01E–08). Details of corresponding terms for clusters 0, 1, and 2 are listed in Supplementary Tables 12, 13, and 14, respectively. The distribution of genes for each term is summarized in Supplementary Fig. 6. a Distribution of significantly enriched terms and gene sets in three clusters of the top 500-proximal gene network. The sunburst diagram (central) shows the proportion of genes involved in each of the three clusters. Each subordinate pie chart (marginal) shows the proportion of genes involved in significantly enriched terms for each cluster. Colors on the subordinate pie charts represent gene types. Solid and stripe patterns indicate functional (participating) and non-functional (non-participating) of the genes in each cluster, respectively. The size of each pie slice corresponds to the number of genes. The proportion is presented as both numbers and percentages. b Categories of significantly enriched terms in three clusters of the top 500-proximal gene network. A sunburst diagram corresponds to a total of 189 significantly enriched terms, comprising of 82, 67, and 40 terms enriched in clusters 0, 1, and 2, respectively. Among six various categories, the Gene Ontology (GO) and Human Phenotype Ontology (HP) were the majority of enrichment in cluster 0; while GO and Human Proteome Atlas (HPA) were the most terms enriched in clusters 1 and 2, respectively. The size of each slice of the pie chart corresponds to the number of enriched terms. Each proportion is displayed as both numbers and percentages. The inner ring shows the proportion of enriched terms for each of the three clusters, colored by clusters. The outer ring shows the proportion of biological term categories for each cluster, colored by categories of enriched terms. c Identification and summarizing features of the top 10 significantly enriched terms of clusters 0, 1, and 2. In brief, cluster 0 had the terms with the greatest enrichment, and almost all the top 10 terms mainly involved in the extracellular matrix (ECM). The top 10 significantly enriched terms in cluster 1 tended to involve in neuronal systems and synapses. Whereas, those in cluster 2 showed various tissue-specific association other than neuronal tissues. Bar plot (left panel) shows the number and types of genes in each of the top 10 enriched terms for each cluster. The x-axis represents the number of genes (gene counts) and gene types distributing to each corresponding term. The y-axis corresponds to significantly enriched terms grouped by clusters, shown in text colors. Colors on bars indicate gene types. Bubble plot (right panel) shows the statistics of enrichment for each of the enriched terms. The size of bubbles represents the number of genes (gene counts) in each corresponding term. The bubble coordinate on the x-axis represents the degree of enrichment for DE genes known as ‘DE gene rich factor’. The DE gene rich factor is the ratio of DE genes in each term to total genes in each term. The larger rich factor represents the greater enrichment. The color scale indicates the degree of significance (B-H q-value ≤ 0.05) in enrichment for each corresponding term (low, dark purple; high, yellow). The significance of enrichment is presented as the -log transformed B-H q-value. For Fig. 3a, c, gene types are indicated by colors: DE, orange; DE/GWAB, light green (a) or yellow (c); GWAB, green; network, purple. B-H Benjamini and Hochberg, DE differentially expressed, GWAB genome-wide association boosting.