Biclustering gene expression datasets - An efficient technique

I. H. Ismail, A. H. Kamal

Research output: A Conference proceeding or a Chapter in BookConference contribution

1 Citation (Scopus)

Abstract

Many techniques have been developed to solve biclustering gene expression datasets problem by minimizing the crossings in the input matrices like cHawk [1] and Bimax [9]. The usage of local searching techniques - in the step of Crossing Minimization (CM) - causes some limitations that affect the accuracy of the obtained biclusters. In this paper, Crossing Minimization Biclustering Algorithm (CMBA) is proposed that deals with this issue. CMBA algorithm consists of two main steps: Crossing Minimization and biclusters Identification. In the former step, two global search techniques: Tabu Search (TS) and Greedy Randomized Adaptive Search Procedure (GRASP) are investigated. While identification step, the Mean Squared Residue (MSR) is used as a measurement for the similarity in the input data. The results show that the CMBA algorithm is competitive to other heuristic techniques. Also, CMBA is able to improve the accuracy for both the low density matrices and the high density matrices. Also, using the MSR approach makes CMBA competitive to well-known C&C algorithm [6].

Original languageEnglish
Title of host publicationINFOS2010 - 2010 7th International Conference on Informatics and Systems
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1-8
Number of pages8
ISBN (Print)9789774033964
Publication statusPublished - 11 Jun 2010
Externally publishedYes
Event2010 7th International Conference on Informatics and Systems, INFOS2010 - Cairo, Egypt
Duration: 28 Mar 201030 Mar 2010

Publication series

NameINFOS2010 - 2010 7th International Conference on Informatics and Systems

Conference

Conference2010 7th International Conference on Informatics and Systems, INFOS2010
CountryEgypt
CityCairo
Period28/03/1030/03/10

Fingerprint Dive into the research topics of 'Biclustering gene expression datasets - An efficient technique'. Together they form a unique fingerprint.

Cite this