A Memetic Approach to Protein Structure Prediction in Triangular Lattices

D. Kamrul Islam; Madhu Chetty; A. Dayem Ullah; Kathleen Steinhofel

A Memetic Approach to Protein Structure Prediction in Triangular Lattices

D. Kamrul Islam, Madhu Chetty, A. Dayem Ullah, Kathleen Steinhofel

Informatics

Monash University

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

1 Citation (Scopus)

Abstract

Protein structure prediction (PSP) remains one of the most challenging open problems in structural bioinformatics. Simplified models in terms of lattice structure and energy function have been proposed to ease the computational hardness of this combinatorial optimization problem. In this paper, we describe a clustered meme-based evolutionary approach for PSP using triangular lattice model. Under the framework of memetic algorithm, the proposed method extracts a pool of cultural information from different regions of the search space using data clustering technique. These highly observed local substructures, termed as meme, are then aggregated centrally for further refinements as second stage of evolution. The optimal utilization of 'explore-and-exploit' feature of evolutionary algorithms is ensured by the inherent parallel architecture of the algorithm and subsequent use of cultural information.

Original language	English
Title of host publication	NEURAL INFORMATION PROCESSING, PT I
Editors	BL Lu, LQ Zhang, J Kwok
Place of Publication	BERLIN
Publisher	Springer
Pages	625-635
Number of pages	11
Volume	7062 LNCS
Edition	PART 1
ISBN (Print)	978-3-642-24955-6
Publication status	Published - 2011
Event	18th International Conference on Neural Information Processing (ICONIP 2011) - Shanghai Duration: 13 Nov 2011 → 17 Nov 2011

Conference

Conference	18th International Conference on Neural Information Processing (ICONIP 2011)
City	Shanghai
Period	13/11/2011 → 17/11/2011

Cite this

@inbook{f34a40517f404071a1b1ec5ed4fe9769,

title = "A Memetic Approach to Protein Structure Prediction in Triangular Lattices",

abstract = "Protein structure prediction (PSP) remains one of the most challenging open problems in structural bioinformatics. Simplified models in terms of lattice structure and energy function have been proposed to ease the computational hardness of this combinatorial optimization problem. In this paper, we describe a clustered meme-based evolutionary approach for PSP using triangular lattice model. Under the framework of memetic algorithm, the proposed method extracts a pool of cultural information from different regions of the search space using data clustering technique. These highly observed local substructures, termed as meme, are then aggregated centrally for further refinements as second stage of evolution. The optimal utilization of 'explore-and-exploit' feature of evolutionary algorithms is ensured by the inherent parallel architecture of the algorithm and subsequent use of cultural information.",

author = "Islam, {D. Kamrul} and Madhu Chetty and Ullah, {A. Dayem} and Kathleen Steinhofel",

year = "2011",

language = "English",

isbn = "978-3-642-24955-6",

volume = "7062 LNCS",

pages = "625--635",

editor = "BL Lu and LQ Zhang and J Kwok",

booktitle = "NEURAL INFORMATION PROCESSING, PT I",

publisher = "Springer",

edition = "PART 1",

note = "18th International Conference on Neural Information Processing (ICONIP 2011) ; Conference date: 13-11-2011 Through 17-11-2011",

}

TY - CHAP

T1 - A Memetic Approach to Protein Structure Prediction in Triangular Lattices

AU - Islam, D. Kamrul

AU - Chetty, Madhu

AU - Ullah, A. Dayem

AU - Steinhofel, Kathleen

PY - 2011

Y1 - 2011

N2 - Protein structure prediction (PSP) remains one of the most challenging open problems in structural bioinformatics. Simplified models in terms of lattice structure and energy function have been proposed to ease the computational hardness of this combinatorial optimization problem. In this paper, we describe a clustered meme-based evolutionary approach for PSP using triangular lattice model. Under the framework of memetic algorithm, the proposed method extracts a pool of cultural information from different regions of the search space using data clustering technique. These highly observed local substructures, termed as meme, are then aggregated centrally for further refinements as second stage of evolution. The optimal utilization of 'explore-and-exploit' feature of evolutionary algorithms is ensured by the inherent parallel architecture of the algorithm and subsequent use of cultural information.

AB - Protein structure prediction (PSP) remains one of the most challenging open problems in structural bioinformatics. Simplified models in terms of lattice structure and energy function have been proposed to ease the computational hardness of this combinatorial optimization problem. In this paper, we describe a clustered meme-based evolutionary approach for PSP using triangular lattice model. Under the framework of memetic algorithm, the proposed method extracts a pool of cultural information from different regions of the search space using data clustering technique. These highly observed local substructures, termed as meme, are then aggregated centrally for further refinements as second stage of evolution. The optimal utilization of 'explore-and-exploit' feature of evolutionary algorithms is ensured by the inherent parallel architecture of the algorithm and subsequent use of cultural information.

M3 - Chapter

SN - 978-3-642-24955-6

VL - 7062 LNCS

SP - 625

EP - 635

BT - NEURAL INFORMATION PROCESSING, PT I

A2 - Lu, BL

A2 - Zhang, LQ

A2 - Kwok, J

PB - Springer

CY - BERLIN

T2 - 18th International Conference on Neural Information Processing (ICONIP 2011)

Y2 - 13 November 2011 through 17 November 2011

ER -

A Memetic Approach to Protein Structure Prediction in Triangular Lattices

Abstract

Conference

Fingerprint

Cite this