Hybrid approach to modeling metabolic systems using genetic algorithm and simplex method

John Yen, David Randolph, Bogju Lee, James C. Liao

Research output: Contribution to journalConference article

9 Citations (Scopus)

Abstract

Genetic algorithms (GA) have been demonstrated to be a promising search and optimization technique that is more likely to converge to a global optimum than most alternative techniques. In an attempt to apply GA to estimate parameters of a metabolic model, however, we found that the slow convergence rate of GA becomes a major problem for its applications to model identification of dynamic systems due to the high computational costs associated with the evaluation of models. To alleviate this difficulty, we developed a hybrid approach that combines Nelder and Mead's downhill simplex method with the genetic algorithm. We evaluated the hybrid approach by extensively comparing its performance with pure GA and pure simplex approaches for the metabolic modeling problem and a function optimization problem. As expected, the hybrid approach not only speeds up GA's rate of convergence, but also improves the quality of the solution found by pure GA.

Original languageEnglish (US)
Pages (from-to)1205-1210
Number of pages6
JournalProceedings of the IEEE International Conference on Systems, Man and Cybernetics
Volume2
StatePublished - Dec 1 1995

Fingerprint

Genetic algorithms
Identification (control systems)
Dynamical systems
Costs

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Hardware and Architecture

Cite this

@article{b3aa550d3903442db7419c46a3590ee0,
title = "Hybrid approach to modeling metabolic systems using genetic algorithm and simplex method",
abstract = "Genetic algorithms (GA) have been demonstrated to be a promising search and optimization technique that is more likely to converge to a global optimum than most alternative techniques. In an attempt to apply GA to estimate parameters of a metabolic model, however, we found that the slow convergence rate of GA becomes a major problem for its applications to model identification of dynamic systems due to the high computational costs associated with the evaluation of models. To alleviate this difficulty, we developed a hybrid approach that combines Nelder and Mead's downhill simplex method with the genetic algorithm. We evaluated the hybrid approach by extensively comparing its performance with pure GA and pure simplex approaches for the metabolic modeling problem and a function optimization problem. As expected, the hybrid approach not only speeds up GA's rate of convergence, but also improves the quality of the solution found by pure GA.",
author = "John Yen and David Randolph and Bogju Lee and Liao, {James C.}",
year = "1995",
month = "12",
day = "1",
language = "English (US)",
volume = "2",
pages = "1205--1210",
journal = "Proceedings of the IEEE International Conference on Systems, Man and Cybernetics",
issn = "0884-3627",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Hybrid approach to modeling metabolic systems using genetic algorithm and simplex method. / Yen, John; Randolph, David; Lee, Bogju; Liao, James C.

In: Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, Vol. 2, 01.12.1995, p. 1205-1210.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Hybrid approach to modeling metabolic systems using genetic algorithm and simplex method

AU - Yen, John

AU - Randolph, David

AU - Lee, Bogju

AU - Liao, James C.

PY - 1995/12/1

Y1 - 1995/12/1

N2 - Genetic algorithms (GA) have been demonstrated to be a promising search and optimization technique that is more likely to converge to a global optimum than most alternative techniques. In an attempt to apply GA to estimate parameters of a metabolic model, however, we found that the slow convergence rate of GA becomes a major problem for its applications to model identification of dynamic systems due to the high computational costs associated with the evaluation of models. To alleviate this difficulty, we developed a hybrid approach that combines Nelder and Mead's downhill simplex method with the genetic algorithm. We evaluated the hybrid approach by extensively comparing its performance with pure GA and pure simplex approaches for the metabolic modeling problem and a function optimization problem. As expected, the hybrid approach not only speeds up GA's rate of convergence, but also improves the quality of the solution found by pure GA.

AB - Genetic algorithms (GA) have been demonstrated to be a promising search and optimization technique that is more likely to converge to a global optimum than most alternative techniques. In an attempt to apply GA to estimate parameters of a metabolic model, however, we found that the slow convergence rate of GA becomes a major problem for its applications to model identification of dynamic systems due to the high computational costs associated with the evaluation of models. To alleviate this difficulty, we developed a hybrid approach that combines Nelder and Mead's downhill simplex method with the genetic algorithm. We evaluated the hybrid approach by extensively comparing its performance with pure GA and pure simplex approaches for the metabolic modeling problem and a function optimization problem. As expected, the hybrid approach not only speeds up GA's rate of convergence, but also improves the quality of the solution found by pure GA.

UR - http://www.scopus.com/inward/record.url?scp=0029507990&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029507990&partnerID=8YFLogxK

M3 - Conference article

VL - 2

SP - 1205

EP - 1210

JO - Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

JF - Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

SN - 0884-3627

ER -