Integrative gene network construction for predicting a set of complementary prostate cancer genes

Ahn, Jaegyoon; Yong-Jin, Yoon; Park, Chihyun; Shin, Eunji; Park, Sang‐Hyun

doi:10.1093/bioinformatics/btr283

Cited by 35 publications

(38 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…31 CD3D was shown to form a complex with many genes, including the cancerrelated genes IKZF1 and TRA@, and activates modification of complexes in the cancer-related genes interleukin 2 (IL2), tyrosine-protein phosphatase non-receptor type 11 (PTPN11), and casitas B-lineage lymphoma (CBL). 32 In conclusion, exome sequencing strategies should be carefully selected and modified depending on what the data set allows. 33 Candidates for implication in ESCC predisposition have been identified after rigorous scrutiny and filteration from the final ESCC variant list.…”

Section: Discussionmentioning

confidence: 99%

Novel candidate genes may be possible predisposing factors revealed by whole exome sequencing in familial esophageal squamous cell carcinoma

et al. 2017

View full text Add to dashboard Cite

Esophageal squamous cell carcinoma is one of the deadliest of all the cancers. Its metastatic properties portend poor prognosis and high rate of recurrence. A more advanced method to identify new molecular biomarkers predicting disease prognosis can be whole exome sequencing. Here, we report the most effective genetic variants of the Notch signaling pathway in esophageal squamous cell carcinoma susceptibility by whole exome sequencing. We analyzed nine probands in unrelated familial esophageal squamous cell carcinoma pedigrees to identify candidate genes. Genomic DNA was extracted and whole exome sequencing performed to generate information about genetic variants in the coding regions. Bioinformatics software applications were utilized to exploit statistical algorithms to demonstrate protein structure and variants conservation. Polymorphic regions were excluded by false-positive investigations. Gene-gene interactions were analyzed for Notch signaling pathway candidates. We identified novel and damaging variants of the Notch signaling pathway through extensive pathway-oriented filtering and functional predictions, which led to the study of 27 candidate novel mutations in all nine patients. Detection of the trinucleotide repeat containing 6B gene mutation (a slice site alteration) in five of the nine probands, but not in any of the healthy samples, suggested that it may be a susceptibility factor for familial esophageal squamous cell carcinoma. Noticeably, 8 of 27 novel candidate gene mutations (e.g. epidermal growth factor, signal transducer and activator of transcription 3, MET) act in a cascade leading to cell survival and proliferation. Our results suggest that the trinucleotide repeat containing 6B mutation may be a candidate predisposing gene in esophageal squamous cell carcinoma. In addition, some of the Notch signaling pathway genetic mutations may act as key contributors to esophageal squamous cell carcinoma.

show abstract

Section: Discussionmentioning

confidence: 99%

Novel candidate genes may be possible predisposing factors revealed by whole exome sequencing in familial esophageal squamous cell carcinoma

et al. 2017

View full text Add to dashboard Cite

show abstract

“…We assumed that each protein in a protein complex also has an interaction that is equal to the interactions of the protein complex. For instance, we can get four binary interactions when a protein complex with four proteins has an interaction with a gene [44]. In total, we used 177,672 interactions for the integrative genetic network, with 15,804 unique proteins.…”

Section: Methodsmentioning

confidence: 99%

A Network-Based Classification Model for Deriving Novel Drug-Disease Associations and Assessing Their Molecular Actions

Ahn

Yong-Jin

2014

PLoS ONE

Self Cite

View full text Add to dashboard Cite

The growing number and variety of genetic network datasets increases the feasibility of understanding how drugs and diseases are associated at the molecular level. Properly selected features of the network representations of existing drug-disease associations can be used to infer novel indications of existing drugs. To find new drug-disease associations, we generated an integrative genetic network using combinations of interactions, including protein-protein interactions and gene regulatory network datasets. Within this network, network adjacencies of drug-drug and disease-disease were quantified using a scored path between target sets of them. Furthermore, the common topological module of drugs or diseases was extracted, and thereby the distance between topological drug-module and disease (or disease-module and drug) was quantified. These quantified scores were used as features for the prediction of novel drug-disease associations. Our classifiers using Random Forest, Multilayer Perceptron and C4.5 showed a high specificity and sensitivity (AUC score of 0.855, 0.828 and 0.797 respectively) in predicting novel drug indications, and displayed a better performance than other methods with limited drug and disease properties. Our predictions and current clinical trials overlap significantly across the different phases of drug development. We also identified and visualized the topological modules of predicted drug indications for certain types of cancers, and for Alzheimer’s disease. Within the network, those modules show potential pathways that illustrate the mechanisms of new drug indications, including propranolol as a potential anticancer agent and telmisartan as treatment for Alzheimer’s disease.

show abstract

“…This database has two classes of interaction data: binary (if two proteins interact directly) and complex (when several proteins form a complex). Since protein complexes also constitute functional groups and take part in activation or inhibition [43] we also included these complexes in our study with the assumption that all proteins in a complex interact with each other. There are a total of 91029 binary interactions between 13494 proteins and 1521 protein complexes that are associated with 3652 proteins.…”

Section: Methodsmentioning

confidence: 99%

Two-layer modular analysis of gene and protein networks in breast cancer

Srivastava

Kumar

Ramaswamy

2014

BMC Systems Biology

View full text Add to dashboard Cite

BackgroundGenomic, proteomic and high-throughput gene expression data, when integrated, can be used to map the interaction networks between genes and proteins. Different approaches have been used to analyze these networks, especially in cancer, where mutations in biologically related genes that encode mutually interacting proteins are believed to be involved. This system of integrated networks as a whole exhibits emergent biological properties that are not obvious at the individual network level. We analyze the system in terms of modules, namely a set of densely interconnected nodes that can be further divided into submodules that are expected to participate in multiple biological activities in coordinated manner.ResultsIn the present work we construct two layers of the breast cancer network: the gene layer, where the correlation network of breast cancer genes is analyzed to identify gene modules, and the protein layer, where each gene module is extended to map out the network of expressed proteins and their interactions in order to identify submodules. Each module and its associated submodules are analyzed to test the robustness of their topological distribution. The constituent biological phenomena are explored through the use of the Gene Ontology. We thus construct a “network of networks”, and demonstrate that both the gene and protein interaction networks are modular in nature. By focusing on the ontological classification, we are able to determine the entire GO profiles that are distributed at different levels of hierarchy. Within each submodule most of the proteins are biologically correlated, and participate in groups of distinct biological activities.ConclusionsThe present approach is an effective method for discovering coherent gene modules and protein submodules. We show that this also provides a means of determining biological pathways (both novel and as well those that have been reported previously) that are related, in the present instance, to breast cancer. Similar strategies are likely to be useful in the analysis of other diseases as well.

show abstract

Integrative gene network construction for predicting a set of complementary prostate cancer genes

Abstract: sanghyun@cs.yonsei.ac.kr

Cited by 35 publications

References 20 publications

Novel candidate genes may be possible predisposing factors revealed by whole exome sequencing in familial esophageal squamous cell carcinoma

Novel candidate genes may be possible predisposing factors revealed by whole exome sequencing in familial esophageal squamous cell carcinoma

A Network-Based Classification Model for Deriving Novel Drug-Disease Associations and Assessing Their Molecular Actions

Two-layer modular analysis of gene and protein networks in breast cancer

Contact Info

Product

Resources

About