Robust experimental design and feature selection in signal transduction pathway modeling

Fei He, Martin Brown, Hong Yue, Lam Fat Yeung

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

2 Citations (Scopus)

Abstract

Due to the general lack of experimental data for biochemical pathway model identification, cell-level time series experimental design is particularly important in current systems biology research. This paper investigates the problem of experimental design for signal transduction pathway modeling, and in particular, focuses on methods for parametric feature selection. An important problem is the estimation of parametric uncertainty which is a function of the true (but unknown) parameters. In this paper, two "robust" feature selection strategies are investigated. The first is a mini-max robust experimental design approach, the second is a sampled experimental design method inspired by the Morris global sensitivity analysis. The two approaches are analyzed and interpreted in terms of a generalized optimal experimental design criterion, and their performance has been compared via simulation on the IκB-NF-κB pathway feature selection problem.

Original languageEnglish
Title of host publication2008 International Joint Conference on Neural Networks, IJCNN 2008
Pages1544-1551
Number of pages8
DOIs
Publication statusPublished - 24 Nov 2008
Externally publishedYes
Event2008 International Joint Conference on Neural Networks, IJCNN 2008 - Hong Kong, China
Duration: 1 Jun 20088 Jun 2008

Conference

Conference2008 International Joint Conference on Neural Networks, IJCNN 2008
CountryChina
CityHong Kong
Period1/06/088/06/08

Fingerprint

Signal transduction
Design of experiments
Feature extraction
Sensitivity analysis
Time series
Identification (control systems)

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

He, F., Brown, M., Yue, H., & Yeung, L. F. (2008). Robust experimental design and feature selection in signal transduction pathway modeling. In 2008 International Joint Conference on Neural Networks, IJCNN 2008 (pp. 1544-1551). [4634001] https://doi.org/10.1109/IJCNN.2008.4634001

Robust experimental design and feature selection in signal transduction pathway modeling. / He, Fei; Brown, Martin; Yue, Hong; Yeung, Lam Fat.

2008 International Joint Conference on Neural Networks, IJCNN 2008. 2008. p. 1544-1551 4634001.

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

He, F, Brown, M, Yue, H & Yeung, LF 2008, Robust experimental design and feature selection in signal transduction pathway modeling. in 2008 International Joint Conference on Neural Networks, IJCNN 2008., 4634001, pp. 1544-1551, 2008 International Joint Conference on Neural Networks, IJCNN 2008, Hong Kong, China, 1/06/08. https://doi.org/10.1109/IJCNN.2008.4634001
He F, Brown M, Yue H, Yeung LF. Robust experimental design and feature selection in signal transduction pathway modeling. In 2008 International Joint Conference on Neural Networks, IJCNN 2008. 2008. p. 1544-1551. 4634001 https://doi.org/10.1109/IJCNN.2008.4634001
He, Fei ; Brown, Martin ; Yue, Hong ; Yeung, Lam Fat. / Robust experimental design and feature selection in signal transduction pathway modeling. 2008 International Joint Conference on Neural Networks, IJCNN 2008. 2008. pp. 1544-1551
@inproceedings{e4b30f9c563d491cb38f54bca99ccd47,
title = "Robust experimental design and feature selection in signal transduction pathway modeling",
abstract = "Due to the general lack of experimental data for biochemical pathway model identification, cell-level time series experimental design is particularly important in current systems biology research. This paper investigates the problem of experimental design for signal transduction pathway modeling, and in particular, focuses on methods for parametric feature selection. An important problem is the estimation of parametric uncertainty which is a function of the true (but unknown) parameters. In this paper, two {"}robust{"} feature selection strategies are investigated. The first is a mini-max robust experimental design approach, the second is a sampled experimental design method inspired by the Morris global sensitivity analysis. The two approaches are analyzed and interpreted in terms of a generalized optimal experimental design criterion, and their performance has been compared via simulation on the IκB-NF-κB pathway feature selection problem.",
author = "Fei He and Martin Brown and Hong Yue and Yeung, {Lam Fat}",
year = "2008",
month = "11",
day = "24",
doi = "10.1109/IJCNN.2008.4634001",
language = "English",
isbn = "9781424418213",
pages = "1544--1551",
booktitle = "2008 International Joint Conference on Neural Networks, IJCNN 2008",

}

TY - GEN

T1 - Robust experimental design and feature selection in signal transduction pathway modeling

AU - He, Fei

AU - Brown, Martin

AU - Yue, Hong

AU - Yeung, Lam Fat

PY - 2008/11/24

Y1 - 2008/11/24

N2 - Due to the general lack of experimental data for biochemical pathway model identification, cell-level time series experimental design is particularly important in current systems biology research. This paper investigates the problem of experimental design for signal transduction pathway modeling, and in particular, focuses on methods for parametric feature selection. An important problem is the estimation of parametric uncertainty which is a function of the true (but unknown) parameters. In this paper, two "robust" feature selection strategies are investigated. The first is a mini-max robust experimental design approach, the second is a sampled experimental design method inspired by the Morris global sensitivity analysis. The two approaches are analyzed and interpreted in terms of a generalized optimal experimental design criterion, and their performance has been compared via simulation on the IκB-NF-κB pathway feature selection problem.

AB - Due to the general lack of experimental data for biochemical pathway model identification, cell-level time series experimental design is particularly important in current systems biology research. This paper investigates the problem of experimental design for signal transduction pathway modeling, and in particular, focuses on methods for parametric feature selection. An important problem is the estimation of parametric uncertainty which is a function of the true (but unknown) parameters. In this paper, two "robust" feature selection strategies are investigated. The first is a mini-max robust experimental design approach, the second is a sampled experimental design method inspired by the Morris global sensitivity analysis. The two approaches are analyzed and interpreted in terms of a generalized optimal experimental design criterion, and their performance has been compared via simulation on the IκB-NF-κB pathway feature selection problem.

UR - http://www.scopus.com/inward/record.url?scp=56349156226&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2008.4634001

DO - 10.1109/IJCNN.2008.4634001

M3 - Conference proceeding

SN - 9781424418213

SP - 1544

EP - 1551

BT - 2008 International Joint Conference on Neural Networks, IJCNN 2008

ER -