Bayesian prognostic model for genomic discovery in bipolar disorder

Swetha S. Bobba, Amin Zollanvari, Gil Alterovitz

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

Abstract

Integrative approaches that incorporate multiple experiments have shown a potential application in the discovery of disease-related attributes. This study presents a unique, data-driven, integrative, Bayesian approach to merge gene expression data from various experiments into prognostic models and evaluate them for the discovery of bipolar-related attributes. Two prognostic models were constructed: a singlystructured Bayesian and a Bayesian multi-net model, which differentiated Bipolar disease state at a higher level of abstraction. These prognostic models were evaluated to find the most common attributes responsible for the disease and their AUROC, using external crossvalidation. The multi-net model achieved an AUROC of 0.907 significantly outperforming the single-structured model with an AUROC of 0.631. The study found six new genes and five chromosomal regions associated with the bipolar state. Enrichment analysis performed in this study revealed biological concepts and proteins responsible for the disease. We anticipate this method and results will be used in the future to integrate information from multiple experiments for the same or related phenotypes of various diseases and also to predict the disease state earlier.

Original languageEnglish
Title of host publicationBIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014
PublisherSciTePress
Pages91-98
Number of pages8
ISBN (Print)9789897580123
Publication statusPublished - 2014
Externally publishedYes
Event5th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2014 - Part of 7th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014 - Angers, Loire Valley, France
Duration: Mar 3 2014Mar 6 2014

Other

Other5th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2014 - Part of 7th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014
CountryFrance
CityAngers, Loire Valley
Period3/3/143/6/14

Fingerprint

Genomics
Disorder
Attribute
Model
Experiment
Experiments
Gene Expression Data
Data-driven
Bayesian Approach
Gene expression
Cross-validation
Phenotype
Genes
Integrate
Gene
Proteins
Protein
Predict
Evaluate

Keywords

  • Bayesian Theory
  • Bipolar Disorder
  • External Cross-Validation
  • Gene Expression

ASJC Scopus subject areas

  • Biomedical Engineering
  • Modelling and Simulation

Cite this

Bobba, S. S., Zollanvari, A., & Alterovitz, G. (2014). Bayesian prognostic model for genomic discovery in bipolar disorder. In BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014 (pp. 91-98). SciTePress.

Bayesian prognostic model for genomic discovery in bipolar disorder. / Bobba, Swetha S.; Zollanvari, Amin; Alterovitz, Gil.

BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014. SciTePress, 2014. p. 91-98.

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

Bobba, SS, Zollanvari, A & Alterovitz, G 2014, Bayesian prognostic model for genomic discovery in bipolar disorder. in BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014. SciTePress, pp. 91-98, 5th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2014 - Part of 7th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014, Angers, Loire Valley, France, 3/3/14.
Bobba SS, Zollanvari A, Alterovitz G. Bayesian prognostic model for genomic discovery in bipolar disorder. In BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014. SciTePress. 2014. p. 91-98
Bobba, Swetha S. ; Zollanvari, Amin ; Alterovitz, Gil. / Bayesian prognostic model for genomic discovery in bipolar disorder. BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014. SciTePress, 2014. pp. 91-98
@inproceedings{ed5f337cc7764542ab66de4a493b66db,
title = "Bayesian prognostic model for genomic discovery in bipolar disorder",
abstract = "Integrative approaches that incorporate multiple experiments have shown a potential application in the discovery of disease-related attributes. This study presents a unique, data-driven, integrative, Bayesian approach to merge gene expression data from various experiments into prognostic models and evaluate them for the discovery of bipolar-related attributes. Two prognostic models were constructed: a singlystructured Bayesian and a Bayesian multi-net model, which differentiated Bipolar disease state at a higher level of abstraction. These prognostic models were evaluated to find the most common attributes responsible for the disease and their AUROC, using external crossvalidation. The multi-net model achieved an AUROC of 0.907 significantly outperforming the single-structured model with an AUROC of 0.631. The study found six new genes and five chromosomal regions associated with the bipolar state. Enrichment analysis performed in this study revealed biological concepts and proteins responsible for the disease. We anticipate this method and results will be used in the future to integrate information from multiple experiments for the same or related phenotypes of various diseases and also to predict the disease state earlier.",
keywords = "Bayesian Theory, Bipolar Disorder, External Cross-Validation, Gene Expression",
author = "Bobba, {Swetha S.} and Amin Zollanvari and Gil Alterovitz",
year = "2014",
language = "English",
isbn = "9789897580123",
pages = "91--98",
booktitle = "BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014",
publisher = "SciTePress",

}

TY - GEN

T1 - Bayesian prognostic model for genomic discovery in bipolar disorder

AU - Bobba, Swetha S.

AU - Zollanvari, Amin

AU - Alterovitz, Gil

PY - 2014

Y1 - 2014

N2 - Integrative approaches that incorporate multiple experiments have shown a potential application in the discovery of disease-related attributes. This study presents a unique, data-driven, integrative, Bayesian approach to merge gene expression data from various experiments into prognostic models and evaluate them for the discovery of bipolar-related attributes. Two prognostic models were constructed: a singlystructured Bayesian and a Bayesian multi-net model, which differentiated Bipolar disease state at a higher level of abstraction. These prognostic models were evaluated to find the most common attributes responsible for the disease and their AUROC, using external crossvalidation. The multi-net model achieved an AUROC of 0.907 significantly outperforming the single-structured model with an AUROC of 0.631. The study found six new genes and five chromosomal regions associated with the bipolar state. Enrichment analysis performed in this study revealed biological concepts and proteins responsible for the disease. We anticipate this method and results will be used in the future to integrate information from multiple experiments for the same or related phenotypes of various diseases and also to predict the disease state earlier.

AB - Integrative approaches that incorporate multiple experiments have shown a potential application in the discovery of disease-related attributes. This study presents a unique, data-driven, integrative, Bayesian approach to merge gene expression data from various experiments into prognostic models and evaluate them for the discovery of bipolar-related attributes. Two prognostic models were constructed: a singlystructured Bayesian and a Bayesian multi-net model, which differentiated Bipolar disease state at a higher level of abstraction. These prognostic models were evaluated to find the most common attributes responsible for the disease and their AUROC, using external crossvalidation. The multi-net model achieved an AUROC of 0.907 significantly outperforming the single-structured model with an AUROC of 0.631. The study found six new genes and five chromosomal regions associated with the bipolar state. Enrichment analysis performed in this study revealed biological concepts and proteins responsible for the disease. We anticipate this method and results will be used in the future to integrate information from multiple experiments for the same or related phenotypes of various diseases and also to predict the disease state earlier.

KW - Bayesian Theory

KW - Bipolar Disorder

KW - External Cross-Validation

KW - Gene Expression

UR - http://www.scopus.com/inward/record.url?scp=84902308433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902308433&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9789897580123

SP - 91

EP - 98

BT - BIOINFORMATICS 2014 - 5th Int. Conf. on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 7th Int. Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2014

PB - SciTePress

ER -