Amoeba-Inspired Heuristic Search Dynamics for Exploring Chemical Reaction Paths

Masashi Aono, Masamitsu Wakabayashi

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We propose a nature-inspired model for simulating chemical reactions in a computationally resource-saving manner. The model was developed by extending our previously proposed heuristic search algorithm, called “AmoebaSAT [Aono et al. 2013],” which was inspired by the spatiotemporal dynamics of a single-celled amoeboid organism that exhibits sophisticated computing capabilities in adapting to its environment efficiently [Zhu et al. 2013]. AmoebaSAT is used for solving an NP-complete combinatorial optimization problem [Garey and Johnson 1979], “the satisfiability problem,” and finds a constraint-satisfying solution at a speed that is dramatically faster than one of the conventionally known fastest stochastic local search methods [Iwama and Tamaki 2004] for a class of randomly generated problem instances [http://www.cs.ubc.ca/~hoos/5/benchm.html]. In cases where the problem has more than one solution, AmoebaSAT exhibits dynamic transition behavior among a variety of the solutions. Inheriting these features of AmoebaSAT, we formulate “AmoebaChem,” which explores a variety of metastable molecules in which several constraints determined by input atoms are satisfied and generates dynamic transition processes among the metastable molecules. AmoebaChem and its developed forms will be applied to the study of the origins of life, to discover reaction paths for which expected or unexpected organic compounds may be formed via unknown unstable intermediates and to estimate the likelihood of each of the discovered paths.

Original languageEnglish
Pages (from-to)339-345
Number of pages7
JournalOrigins of Life and Evolution of Biospheres
Volume45
Issue number3
DOIs
Publication statusPublished - 2015 Sep 25
Externally publishedYes

Fingerprint

amoeba
Amoeba
chemical reactions
heuristics
chemical reaction
origin of life
organic compound
system optimization
organic compounds
organisms
molecules
resources
resource
optimization
estimates
atoms
methodology

Keywords

  • Chemical reaction
  • Combinatorial optimization
  • Constraint satisfaction
  • Metastable states
  • Octet rule
  • Spatiotemporal dynamics

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Space and Planetary Science

Cite this

Amoeba-Inspired Heuristic Search Dynamics for Exploring Chemical Reaction Paths. / Aono, Masashi; Wakabayashi, Masamitsu.

In: Origins of Life and Evolution of Biospheres, Vol. 45, No. 3, 25.09.2015, p. 339-345.

Research output: Contribution to journalArticle

@article{d7a4b20f6c074ae09fea2c0d7470ac6c,
title = "Amoeba-Inspired Heuristic Search Dynamics for Exploring Chemical Reaction Paths",
abstract = "We propose a nature-inspired model for simulating chemical reactions in a computationally resource-saving manner. The model was developed by extending our previously proposed heuristic search algorithm, called “AmoebaSAT [Aono et al. 2013],” which was inspired by the spatiotemporal dynamics of a single-celled amoeboid organism that exhibits sophisticated computing capabilities in adapting to its environment efficiently [Zhu et al. 2013]. AmoebaSAT is used for solving an NP-complete combinatorial optimization problem [Garey and Johnson 1979], “the satisfiability problem,” and finds a constraint-satisfying solution at a speed that is dramatically faster than one of the conventionally known fastest stochastic local search methods [Iwama and Tamaki 2004] for a class of randomly generated problem instances [http://www.cs.ubc.ca/~hoos/5/benchm.html]. In cases where the problem has more than one solution, AmoebaSAT exhibits dynamic transition behavior among a variety of the solutions. Inheriting these features of AmoebaSAT, we formulate “AmoebaChem,” which explores a variety of metastable molecules in which several constraints determined by input atoms are satisfied and generates dynamic transition processes among the metastable molecules. AmoebaChem and its developed forms will be applied to the study of the origins of life, to discover reaction paths for which expected or unexpected organic compounds may be formed via unknown unstable intermediates and to estimate the likelihood of each of the discovered paths.",
keywords = "Chemical reaction, Combinatorial optimization, Constraint satisfaction, Metastable states, Octet rule, Spatiotemporal dynamics",
author = "Masashi Aono and Masamitsu Wakabayashi",
year = "2015",
month = "9",
day = "25",
doi = "10.1007/s11084-015-9450-5",
language = "English",
volume = "45",
pages = "339--345",
journal = "Origins of Life and Evolution of the Biosphere",
issn = "0169-6149",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Amoeba-Inspired Heuristic Search Dynamics for Exploring Chemical Reaction Paths

AU - Aono, Masashi

AU - Wakabayashi, Masamitsu

PY - 2015/9/25

Y1 - 2015/9/25

N2 - We propose a nature-inspired model for simulating chemical reactions in a computationally resource-saving manner. The model was developed by extending our previously proposed heuristic search algorithm, called “AmoebaSAT [Aono et al. 2013],” which was inspired by the spatiotemporal dynamics of a single-celled amoeboid organism that exhibits sophisticated computing capabilities in adapting to its environment efficiently [Zhu et al. 2013]. AmoebaSAT is used for solving an NP-complete combinatorial optimization problem [Garey and Johnson 1979], “the satisfiability problem,” and finds a constraint-satisfying solution at a speed that is dramatically faster than one of the conventionally known fastest stochastic local search methods [Iwama and Tamaki 2004] for a class of randomly generated problem instances [http://www.cs.ubc.ca/~hoos/5/benchm.html]. In cases where the problem has more than one solution, AmoebaSAT exhibits dynamic transition behavior among a variety of the solutions. Inheriting these features of AmoebaSAT, we formulate “AmoebaChem,” which explores a variety of metastable molecules in which several constraints determined by input atoms are satisfied and generates dynamic transition processes among the metastable molecules. AmoebaChem and its developed forms will be applied to the study of the origins of life, to discover reaction paths for which expected or unexpected organic compounds may be formed via unknown unstable intermediates and to estimate the likelihood of each of the discovered paths.

AB - We propose a nature-inspired model for simulating chemical reactions in a computationally resource-saving manner. The model was developed by extending our previously proposed heuristic search algorithm, called “AmoebaSAT [Aono et al. 2013],” which was inspired by the spatiotemporal dynamics of a single-celled amoeboid organism that exhibits sophisticated computing capabilities in adapting to its environment efficiently [Zhu et al. 2013]. AmoebaSAT is used for solving an NP-complete combinatorial optimization problem [Garey and Johnson 1979], “the satisfiability problem,” and finds a constraint-satisfying solution at a speed that is dramatically faster than one of the conventionally known fastest stochastic local search methods [Iwama and Tamaki 2004] for a class of randomly generated problem instances [http://www.cs.ubc.ca/~hoos/5/benchm.html]. In cases where the problem has more than one solution, AmoebaSAT exhibits dynamic transition behavior among a variety of the solutions. Inheriting these features of AmoebaSAT, we formulate “AmoebaChem,” which explores a variety of metastable molecules in which several constraints determined by input atoms are satisfied and generates dynamic transition processes among the metastable molecules. AmoebaChem and its developed forms will be applied to the study of the origins of life, to discover reaction paths for which expected or unexpected organic compounds may be formed via unknown unstable intermediates and to estimate the likelihood of each of the discovered paths.

KW - Chemical reaction

KW - Combinatorial optimization

KW - Constraint satisfaction

KW - Metastable states

KW - Octet rule

KW - Spatiotemporal dynamics

UR - http://www.scopus.com/inward/record.url?scp=84938952462&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84938952462&partnerID=8YFLogxK

U2 - 10.1007/s11084-015-9450-5

DO - 10.1007/s11084-015-9450-5

M3 - Article

C2 - 26129639

AN - SCOPUS:84938952462

VL - 45

SP - 339

EP - 345

JO - Origins of Life and Evolution of the Biosphere

JF - Origins of Life and Evolution of the Biosphere

SN - 0169-6149

IS - 3

ER -