Limits...
Network perturbation by recurrent regulatory variants in cancer

View Article: PubMed Central - PubMed

ABSTRACT

Cancer driving genes have been identified as recurrently affected by variants that alter protein-coding sequences. However, a majority of cancer variants arise in noncoding regions, and some of them are thought to play a critical role through transcriptional perturbation. Here we identified putative transcriptional driver genes based on combinatorial variant recurrence in cis-regulatory regions. The identified genes showed high connectivity in the cancer type-specific transcription regulatory network, with high outdegree and many downstream genes, highlighting their causative role during tumorigenesis. In the protein interactome, the identified transcriptional drivers were not as highly connected as coding driver genes but appeared to form a network module centered on the coding drivers. The coding and regulatory variants associated via these interactions between the coding and transcriptional drivers showed exclusive and complementary occurrence patterns across tumor samples. Transcriptional cancer drivers may act through an extensive perturbation of the regulatory network and by altering protein network modules through interactions with coding driver genes.

No MeSH data available.


Related in: MedlinePlus

Combinatorial cis-regulatory recurrence.(A) Illustration of our recurrence model. Four variants from different samples are scattered in cis-regulatory regions but converge on the same gene via chromatin interactions. (B) A radar plot showing the significance of enrichment for eight cancer-related Gene Ontology terms. The length of the plot scales with log10 (P value). The P values were derived from the hypergeometric distribution and adjusted for multiple testing by the Bonferroni correction. (C) Relative causal score of the TDs grouped by the recurrence level and the CDs (CGC and 20/20) in the Bayesian network of breast cancer. Causal scores were calculated as described in the Methods and normalized by dividing by the average causal score of all genes in the network. (D) The relative degree of the TDs and CDs in the coexpression network in breast cancer. The degree was divided by the network average. (E) Schematic illustration of genomic simulation (in silico or clinical) in which variants are randomized, and epigenomic simulation in which K562 chromatin interactome is used in place of MCF-7 and HepG2.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5383347&req=5

pcbi.1005449.g001: Combinatorial cis-regulatory recurrence.(A) Illustration of our recurrence model. Four variants from different samples are scattered in cis-regulatory regions but converge on the same gene via chromatin interactions. (B) A radar plot showing the significance of enrichment for eight cancer-related Gene Ontology terms. The length of the plot scales with log10 (P value). The P values were derived from the hypergeometric distribution and adjusted for multiple testing by the Bonferroni correction. (C) Relative causal score of the TDs grouped by the recurrence level and the CDs (CGC and 20/20) in the Bayesian network of breast cancer. Causal scores were calculated as described in the Methods and normalized by dividing by the average causal score of all genes in the network. (D) The relative degree of the TDs and CDs in the coexpression network in breast cancer. The degree was divided by the network average. (E) Schematic illustration of genomic simulation (in silico or clinical) in which variants are randomized, and epigenomic simulation in which K562 chromatin interactome is used in place of MCF-7 and HepG2.

Mentions: The workflow of our analyses is summarized in S1 Fig. We first identified regulatory variants in 119 breast and 88 liver cancer samples as illustrated in Fig 1A. In this example, four different samples carry motif-changing variants at different positions in cis-regulatory regions, whose convergence on a common transcriptional target is revealed by the chromatin interactome. In this case, the combinatorial measure of variant recurrence for this gene should be four although none of the four variants arose at the same site. For this type of recurrence analysis, we employed enhancer-promoter maps constructed by RNA polymerase II-mediated chromatin interaction analysis by paired-end tag (ChIA-PET) sequencing [26–28], integrated methods for predicting enhancer targets (IM-PET) [29], DHS tag density correlations [1], and cap analysis gene expression (CAGE)-based RNA correlations [30]. We also applied additional filters for enhancer-promoter mapping (see Methods). The different criteria and resulting number of chromatin interactions are described in S2 Fig. The list of genes with the resulting recurrence level in each cancer is provided in S1 Table.


Network perturbation by recurrent regulatory variants in cancer
Combinatorial cis-regulatory recurrence.(A) Illustration of our recurrence model. Four variants from different samples are scattered in cis-regulatory regions but converge on the same gene via chromatin interactions. (B) A radar plot showing the significance of enrichment for eight cancer-related Gene Ontology terms. The length of the plot scales with log10 (P value). The P values were derived from the hypergeometric distribution and adjusted for multiple testing by the Bonferroni correction. (C) Relative causal score of the TDs grouped by the recurrence level and the CDs (CGC and 20/20) in the Bayesian network of breast cancer. Causal scores were calculated as described in the Methods and normalized by dividing by the average causal score of all genes in the network. (D) The relative degree of the TDs and CDs in the coexpression network in breast cancer. The degree was divided by the network average. (E) Schematic illustration of genomic simulation (in silico or clinical) in which variants are randomized, and epigenomic simulation in which K562 chromatin interactome is used in place of MCF-7 and HepG2.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5383347&req=5

pcbi.1005449.g001: Combinatorial cis-regulatory recurrence.(A) Illustration of our recurrence model. Four variants from different samples are scattered in cis-regulatory regions but converge on the same gene via chromatin interactions. (B) A radar plot showing the significance of enrichment for eight cancer-related Gene Ontology terms. The length of the plot scales with log10 (P value). The P values were derived from the hypergeometric distribution and adjusted for multiple testing by the Bonferroni correction. (C) Relative causal score of the TDs grouped by the recurrence level and the CDs (CGC and 20/20) in the Bayesian network of breast cancer. Causal scores were calculated as described in the Methods and normalized by dividing by the average causal score of all genes in the network. (D) The relative degree of the TDs and CDs in the coexpression network in breast cancer. The degree was divided by the network average. (E) Schematic illustration of genomic simulation (in silico or clinical) in which variants are randomized, and epigenomic simulation in which K562 chromatin interactome is used in place of MCF-7 and HepG2.
Mentions: The workflow of our analyses is summarized in S1 Fig. We first identified regulatory variants in 119 breast and 88 liver cancer samples as illustrated in Fig 1A. In this example, four different samples carry motif-changing variants at different positions in cis-regulatory regions, whose convergence on a common transcriptional target is revealed by the chromatin interactome. In this case, the combinatorial measure of variant recurrence for this gene should be four although none of the four variants arose at the same site. For this type of recurrence analysis, we employed enhancer-promoter maps constructed by RNA polymerase II-mediated chromatin interaction analysis by paired-end tag (ChIA-PET) sequencing [26–28], integrated methods for predicting enhancer targets (IM-PET) [29], DHS tag density correlations [1], and cap analysis gene expression (CAGE)-based RNA correlations [30]. We also applied additional filters for enhancer-promoter mapping (see Methods). The different criteria and resulting number of chromatin interactions are described in S2 Fig. The list of genes with the resulting recurrence level in each cancer is provided in S1 Table.

View Article: PubMed Central - PubMed

ABSTRACT

Cancer driving genes have been identified as recurrently affected by variants that alter protein-coding sequences. However, a majority of cancer variants arise in noncoding regions, and some of them are thought to play a critical role through transcriptional perturbation. Here we identified putative transcriptional driver genes based on combinatorial variant recurrence in cis-regulatory regions. The identified genes showed high connectivity in the cancer type-specific transcription regulatory network, with high outdegree and many downstream genes, highlighting their causative role during tumorigenesis. In the protein interactome, the identified transcriptional drivers were not as highly connected as coding driver genes but appeared to form a network module centered on the coding drivers. The coding and regulatory variants associated via these interactions between the coding and transcriptional drivers showed exclusive and complementary occurrence patterns across tumor samples. Transcriptional cancer drivers may act through an extensive perturbation of the regulatory network and by altering protein network modules through interactions with coding driver genes.

No MeSH data available.


Related in: MedlinePlus