Improving biomolecular pattern discovery and visualization with hybrid self-adaptive networks

Author

Wang, Haiying ; Azuaje, Francisco ; Black, Norman

Author_Institution

Sch. of Comput. & Math., Univ. of Ulster, Jordanstown, UK

Volume

1

Issue

4

fYear

2002

Firstpage

146

Lastpage

166

Abstract

There is an increasing need to develop powerful techniques to improve biomedical pattern discovery and visualization. This paper presents an automated approach, based on hybrid self-adaptive neural networks, to pattern identification and visualization for biomolecular data. The methods are tested on two datasets: leukemia expression data and DNA splice-junction sequences. Several supervised and unsupervised models are implemented and compared. A comprehensive evaluation study of some of their intrinsic mechanisms is presented. The results suggest that these tools may be useful to support biological knowledge discovery based on advanced classification and visualization tasks.

Keywords

DNA; biology computing; data mining; genetics; molecular biophysics; pattern classification; pattern clustering; self-organising feature maps; unsupervised learning; DNA splice-junction sequences; advanced classification tasks; automated approach; biological knowledge discovery; biomolecular pattern discovery; biomolecular visualization; hybrid self-adaptive neural networks; intrinsic mechanisms; leukemia expression data; pattern identification; supervised models; unsupervised models; Application software; Bioinformatics; Clustering algorithms; DNA; Data mining; Data visualization; Genetics; Neural networks; Pattern analysis; Sequences;

fLanguage

English

Journal_Title

NanoBioscience, IEEE Transactions on

Publisher

ieee

ISSN

1536-1241

Type

jour

DOI

10.1109/TNB.2003.809465

Filename

1195403