Functional data analysis of the dynamics of gene regulatory networks

Tomohiro Ando, Seiya Imoto, Satoru Miyano

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

4 Citations (Scopus)

Abstract

A new method for constructing gene networks from microarray time-series gene expression data is proposed in the context of Bayesian network approach. An essential point of Bayesian network modeling is the construction of the conditional distribution of each random variable. When estimating the conditional distributions from gene expression data, a common problem is that gene expression data contain multiple missing values. Unfortunately, many methods for constructing conditional distributions require a complete gene expression value and may lose effectiveness even with a few missing value. Additionally, they treat microarray time-series gene expression data as static data, although time can be an important factor that affects the gene expression levels. We overcome these difficulties by using the method of functional data analysis. The proposed network construction method consists of two stages. Firstly, discrete microarray time-series gene expression values are expressed as a continuous curve of time. To account for the time dependency of gene expression measurements and the noisy nature of the microarray data, P-spline nonlinear regression models are utilized. After this preprocessing step, the conditional distribution of each random variable is constructed based on functional linear regression models. The effectiveness of the proposed method is investigated through Monte Carlo simulations and the analysis of Saccharomyces cerevisiae gene expression data.

Original languageEnglish
Title of host publicationLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
EditorsJ.A. Lopez, E. Benfenati, W. Dubitzky
Pages69-83
Number of pages15
Volume3303
Publication statusPublished - 2004
Externally publishedYes
EventInternational Symposium KELSI 2004: Knowledge Exploration in Life Science Informatics - Milan, Italy
Duration: 2004 Nov 252004 Nov 26

Other

OtherInternational Symposium KELSI 2004: Knowledge Exploration in Life Science Informatics
CountryItaly
CityMilan
Period04/11/2504/11/26

Fingerprint

Gene expression
Genes
Microarrays
Time series
Bayesian networks
Random variables
Linear regression
Splines
Yeast

Keywords

  • Bayesian networks
  • Functional data analysis
  • P-spline
  • Smoothing
  • Time-series gene expression data

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Ando, T., Imoto, S., & Miyano, S. (2004). Functional data analysis of the dynamics of gene regulatory networks. In J. A. Lopez, E. Benfenati, & W. Dubitzky (Eds.), Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3303, pp. 69-83)

Functional data analysis of the dynamics of gene regulatory networks. / Ando, Tomohiro; Imoto, Seiya; Miyano, Satoru.

Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). ed. / J.A. Lopez; E. Benfenati; W. Dubitzky. Vol. 3303 2004. p. 69-83.

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

Ando, T, Imoto, S & Miyano, S 2004, Functional data analysis of the dynamics of gene regulatory networks. in JA Lopez, E Benfenati & W Dubitzky (eds), Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). vol. 3303, pp. 69-83, International Symposium KELSI 2004: Knowledge Exploration in Life Science Informatics, Milan, Italy, 04/11/25.
Ando T, Imoto S, Miyano S. Functional data analysis of the dynamics of gene regulatory networks. In Lopez JA, Benfenati E, Dubitzky W, editors, Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). Vol. 3303. 2004. p. 69-83
Ando, Tomohiro ; Imoto, Seiya ; Miyano, Satoru. / Functional data analysis of the dynamics of gene regulatory networks. Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). editor / J.A. Lopez ; E. Benfenati ; W. Dubitzky. Vol. 3303 2004. pp. 69-83
@inproceedings{da8b825c70de41aabac23d35f76421bd,
title = "Functional data analysis of the dynamics of gene regulatory networks",
abstract = "A new method for constructing gene networks from microarray time-series gene expression data is proposed in the context of Bayesian network approach. An essential point of Bayesian network modeling is the construction of the conditional distribution of each random variable. When estimating the conditional distributions from gene expression data, a common problem is that gene expression data contain multiple missing values. Unfortunately, many methods for constructing conditional distributions require a complete gene expression value and may lose effectiveness even with a few missing value. Additionally, they treat microarray time-series gene expression data as static data, although time can be an important factor that affects the gene expression levels. We overcome these difficulties by using the method of functional data analysis. The proposed network construction method consists of two stages. Firstly, discrete microarray time-series gene expression values are expressed as a continuous curve of time. To account for the time dependency of gene expression measurements and the noisy nature of the microarray data, P-spline nonlinear regression models are utilized. After this preprocessing step, the conditional distribution of each random variable is constructed based on functional linear regression models. The effectiveness of the proposed method is investigated through Monte Carlo simulations and the analysis of Saccharomyces cerevisiae gene expression data.",
keywords = "Bayesian networks, Functional data analysis, P-spline, Smoothing, Time-series gene expression data",
author = "Tomohiro Ando and Seiya Imoto and Satoru Miyano",
year = "2004",
language = "English",
volume = "3303",
pages = "69--83",
editor = "J.A. Lopez and E. Benfenati and W. Dubitzky",
booktitle = "Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)",

}

TY - GEN

T1 - Functional data analysis of the dynamics of gene regulatory networks

AU - Ando, Tomohiro

AU - Imoto, Seiya

AU - Miyano, Satoru

PY - 2004

Y1 - 2004

N2 - A new method for constructing gene networks from microarray time-series gene expression data is proposed in the context of Bayesian network approach. An essential point of Bayesian network modeling is the construction of the conditional distribution of each random variable. When estimating the conditional distributions from gene expression data, a common problem is that gene expression data contain multiple missing values. Unfortunately, many methods for constructing conditional distributions require a complete gene expression value and may lose effectiveness even with a few missing value. Additionally, they treat microarray time-series gene expression data as static data, although time can be an important factor that affects the gene expression levels. We overcome these difficulties by using the method of functional data analysis. The proposed network construction method consists of two stages. Firstly, discrete microarray time-series gene expression values are expressed as a continuous curve of time. To account for the time dependency of gene expression measurements and the noisy nature of the microarray data, P-spline nonlinear regression models are utilized. After this preprocessing step, the conditional distribution of each random variable is constructed based on functional linear regression models. The effectiveness of the proposed method is investigated through Monte Carlo simulations and the analysis of Saccharomyces cerevisiae gene expression data.

AB - A new method for constructing gene networks from microarray time-series gene expression data is proposed in the context of Bayesian network approach. An essential point of Bayesian network modeling is the construction of the conditional distribution of each random variable. When estimating the conditional distributions from gene expression data, a common problem is that gene expression data contain multiple missing values. Unfortunately, many methods for constructing conditional distributions require a complete gene expression value and may lose effectiveness even with a few missing value. Additionally, they treat microarray time-series gene expression data as static data, although time can be an important factor that affects the gene expression levels. We overcome these difficulties by using the method of functional data analysis. The proposed network construction method consists of two stages. Firstly, discrete microarray time-series gene expression values are expressed as a continuous curve of time. To account for the time dependency of gene expression measurements and the noisy nature of the microarray data, P-spline nonlinear regression models are utilized. After this preprocessing step, the conditional distribution of each random variable is constructed based on functional linear regression models. The effectiveness of the proposed method is investigated through Monte Carlo simulations and the analysis of Saccharomyces cerevisiae gene expression data.

KW - Bayesian networks

KW - Functional data analysis

KW - P-spline

KW - Smoothing

KW - Time-series gene expression data

UR - http://www.scopus.com/inward/record.url?scp=22944434182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=22944434182&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:22944434182

VL - 3303

SP - 69

EP - 83

BT - Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)

A2 - Lopez, J.A.

A2 - Benfenati, E.

A2 - Dubitzky, W.

ER -