Detecting gene interactions within a Bayesian Network framework using external knowledge

Senol Isci, Umut Agyuz, Cengizhan Ozturk, Hasan H. Otu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

Biological and clinical databases are increasing at a very high rate making a large volume of experimental data publicly available. In this paper, we propose a framework that makes use of external biological knowledge to predict if two given genes interact with each other. To this end, we utilize prior knowledge about interaction of two genes by generating a Bayesian Network using existing external biological knowledge. External knowledge types to be utilized are obtained from interaction databases such as BioGrid and Reac-tome and consist of protein-protein, protein-DNA/RNA, and gene interactions. We first built a naïve Bayesian Network to predict if two genes interact by employing parameter learning using known gene interactions. We propose that the resulting model will be incorporated into methods learning networks from high throughput biological data and interacting genes will be represented in the form of a network. In this process of network generation, the Bayesian Network model deducing gene interactions from external knowledge will be used to calculate the probability of candidate networks to enhance the structure learning task. Our simulations on both synthetic and real data sets show that proposed framework can successfully enhance identification of the true network and be used in predicting gene interactions.

Original languageEnglish (US)
Title of host publication2012 7th International Symposium on Health Informatics and Bioinformatics, HIBIT 2012
Pages82-87
Number of pages6
DOIs
StatePublished - 2012
Externally publishedYes
Event2012 7th International Symposium on Health Informatics and Bioinformatics, HIBIT 2012 - Cappadocia, Turkey
Duration: Apr 19 2012Apr 22 2012

Publication series

Name2012 7th International Symposium on Health Informatics and Bioinformatics, HIBIT 2012

Other

Other2012 7th International Symposium on Health Informatics and Bioinformatics, HIBIT 2012
Country/TerritoryTurkey
CityCappadocia
Period4/19/124/22/12

ASJC Scopus subject areas

  • Health Informatics
  • Health Information Management

Fingerprint

Dive into the research topics of 'Detecting gene interactions within a Bayesian Network framework using external knowledge'. Together they form a unique fingerprint.

Cite this