We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Results: 5

1.
Figure 5

Figure 5. BioPAX to binary interaction mapping: example mapping rule.. From: Automated Network Analysis Identifies Core Pathways in Glioblastoma.

To integrate complex signaling pathway data into our Human Interaction Network (HIN), we have developed a set of rules for mapping subgraphs of biochemical networks to binary interactions. An example rule for mapping state changes, such as phosphorylation events, is shown.

Ethan Cerami, et al. PLoS One. 2010;5(2):e8918.
2.
Figure 2

Figure 2. Network modules identified in GBM.. From: Automated Network Analysis Identifies Core Pathways in Glioblastoma.

(A) Modules are densely connected sets of altered genes that may reflect oncogenic processes. A total of 10 modules were identified, the largest of which are shown. Linker genes, indicated in red, are not altered in GBM, but are statistically enriched for connections to GBM-altered genes. (B) The observed modularity of the GBM network (0.519) is compared with 1000 randomly rewired networks (average 0.296, standard deviation 0.058). This results in a z-score, or scaled modularity score, of 3.84.

Ethan Cerami, et al. PLoS One. 2010;5(2):e8918.
3.
Figure 4

Figure 4. Network analysis identifies three additional altered modules, including the DCTN2 module, which is involved in microtubule organization.. From: Automated Network Analysis Identifies Core Pathways in Glioblastoma.

Each of the altered modules is implicated by homozygous deletions or multi-copy amplifications across the 84 analyzed GBM cases. Each module is annotated with Gene Ontology enrichment, chromosome location, statistical significance of copy number alteration against a background model of random aberrations, as determined by RAE copy number analysis; assessment of correlation between copy number and mRNA expression, and genomic signature across 84 GBM cases.

Ethan Cerami, et al. PLoS One. 2010;5(2):e8918.
4.
Figure 1

Figure 1. Overview of the network approach for identifying oncogenic processes and candidate driver genes in GBM.. From: Automated Network Analysis Identifies Core Pathways in Glioblastoma.

We began by creating a literature curated Human Interaction Network (HIN) derived from protein-protein interactions and signaling pathways (A), and assembling genomic alterations in GBM (B). We then extracted a GBM-specific network of altered genes (C), which was then partitioned into network modules (D). We assessed the level of connectivity seen within the GBM network by using (E1) a global null model to compare the size of the largest component in the observed network v. networks arising from randomly selected gene sets; and (E2) a local null model to compare network modularity of the observed network to locally rewired networks.

Ethan Cerami, et al. PLoS One. 2010;5(2):e8918.
5.
Figure 3

Figure 3. Automated network analysis approach is in close agreement with previous manually curated pathway analysis approach.. From: Automated Network Analysis Identifies Core Pathways in Glioblastoma.

The original pathway analysis of TCGA glioblastoma datasets was derived by mapping observed gene alterations onto a manually curated GBM-specific network, based on the glioblastoma literature. This non-algorithmic analysis identified driver alterations in the p53, RB and PI3K pathways. Our automated network analysis approach is in close agreement with these results (top: P53/Rb; bottom: PI3K). The one main exception is that network analysis does not identify NF1 as a participant in the PI3K module. Additional candidate driver genes identified by network analysis, including AGAP2, are identified and annotated on the right. Percentage values after each newly identified candidate driver indicate percent of cases with genetic alterations (sequence mutations, homozygous deletions, or multi-copy amplifications) across the 84 TCGA GBM cases analyzed.

Ethan Cerami, et al. PLoS One. 2010;5(2):e8918.

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Write to the Help Desk