Robust experimental design and feature selection in signal transduction pathway modeling

F. He, M. Brown, H. Yue, L.F. Yeung

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

2 Citations (Scopus)


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 ldquorobustrdquo 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 IkappaB-NF-kappaB pathway feature selection problem.
Original languageEnglish
Title of host publicationIEEE International Joint Conference on Neural Networks, 2008. IJCNN 2008.
Place of PublicationHong Kong, China
Number of pages8
ISBN (Print)978-1-4244-1820-6
Publication statusPublished - Jun 2008
Event2008 IEEE International Joint Conference on Neural Networks - Hong Kong, China
Duration: 1 Jun 20088 Jun 2008


Conference2008 IEEE International Joint Conference on Neural Networks
CityHong Kong


  • biochemistry
  • design of experiments
  • feature extraction
  • sensitivity analysis
  • Morris global sensitivity analysis
  • biochemical pathway model identification
  • feature selection
  • robust experimental design
  • signal transduction pathway modeling


Dive into the research topics of 'Robust experimental design and feature selection in signal transduction pathway modeling'. Together they form a unique fingerprint.

Cite this