Fig. 2: Summary results of the top 500- and top 2000-proximal gene networks. | Molecular Psychiatry

Fig. 2: Summary results of the top 500- and top 2000-proximal gene networks.

From: Focal adhesion is associated with lithium response in bipolar disorder: evidence from a network-based multi-omics analysis

Fig. 2

a Results of hypergeometric test between the top 500 or top 2000 proximal network genes and the top 5% prioritized genes. A line graph (left panel) plots the -log (hypergeometric P-value; y-axis) against the number of proximal network genes (x-axis). Each line corresponds to the top 5% prioritized gene sets for GWAB [31] (orange) and VEGAS [30] (green). Vertical dashed grey lines at 500 and 2000 (on the x-axis) represent the top 500 and top 2000 proximal network genes, respectively. A table (right panel) shows summary of hypergeometric test results. A hypergeometric P-value of < 0.05 shows the significance of overlap in genes between the top 500 or top 2000 proximal network genes and the top 5% prioritized genes obtained by either GWAB (1119 genes) or VEGAS (1180 genes). See the lists of the top 5% prioritized genes and the top 500 and top 2000 proximal network genes in Supplementary Tables 9 and 10, respectively. See the detailed results of the hypergeometric test in Supplementary Table 11. b Distribution of genes in the top 500- and top 2000-proximal gene networks derived from the 41 protein-coding DE genes. Pie charts show the number and types of genes in the proximal gene networks: the top 500-proximal gene network (left) and the top 2000-proximal gene network (right). Each pie chart represents the proportion of gene types as both numbers and percentages. Colors represent gene types (DE, orange; DE/GWAB, yellow; GWAB, green; network, purple). The list of 41 protein-coding DE genes is presented in Fig. 1c, d, and Supplementary Table 8. c, d Network propagation of the top 500 and top 2000 proximal network genes. The network used 41 protein-coding DE genes as the seeds and GIANT interactome [37] (Tissue-specific gene networks from HumanBase; https://hb.flatironinstitute.org/about) as the background. Color scale on genes (nodes) represents the degree of proximity from the seeds (low/further, yellow; high/nearer, red). Shapes represent gene types. c Network propagation of the top 500 proximal network genes. The network resulted in 500 genes (nodes) connected with 25 025 edges, and contained 30 DE, 4 DE/GWAB, 69 GWAB, and 397 network genes. A subset of 103 (69 GWAB, 4 GWAB/seed, and 30 seed) genes showed statistical significance of the overlap (Phypergeometric = 1.28E–09) in genes between the top 500-proximal gene network and top 5% GWAB-prioritized data, two independent sources of data. The interactive graph of the top 500-proximal gene network can be accessed at https://www.ndexbio.org/viewer/networks/bc0c278d-aea2-11eb-9e72-0ac135e8bacf. d Network propagation of the top 2000 proximal network genes. The network resulted in 2000 genes (nodes) connected with 157 688 edges, and contained 30 DE, 4 DE/GWAB, 237 GWAB, and 1729 network genes. Note that a subset of 271 (237 GWAB, 4 GWAB/seed, and 30 seed) genes with 3135 edges are shown here. These relevant 271 genes showed statistically significant overlap (Phypergeometric = 4.10E–18) between the top 2000-proximal gene network and top 5% GWAB-prioritized data. The interactive graph of the top 2000-proximal gene network can be accessed at https://ndexbio.org/viewer/networks/8ddb8cc6-aea3–11eb-9e72-0ac135e8bacf. e Cluster enrichment analysis of the top 500-proximal gene network derived from the 41 protein-coding DE genes. Three clusters were identified based on biological functions. Colors on genes (nodes) indicate for each cluster (0, blue; 1, orange; 2, green). Shapes represent gene types. The basic network prior cluster enrichment analysis is displayed in Fig. 2c. See the list and distribution of genes for each cluster in Supplementary Table 10 and Fig. 3, respectively. The interactive graph of cluster enrichment for the top 500-proximal gene network can be publicly accessed at https://ndexbio.org/viewer/networks/43073550-aea3-11eb-9e72-0ac135e8bacf. For Fig. 2c–e, gene types are indicated by shapes: DE, filled triangle; DE/GWAB, filled outlined triangle; GWAB, filled outlined square; network, filled circle. DE differentially expressed, GWAB genome-wide association boosting, VEGAS versatile gene-based association study.

Back to article page