Making genome expression data meaningful: prediction and discovery of classes of cancer through a connectionist learning approach

Author

Azuaje, Francisco

Author_Institution

Dept. of Comput. Sci., Trinity Coll., Dublin, Ireland

fYear

2000

fDate

2000

Firstpage

208

Lastpage

213

Abstract

Despite more than 30 years of experimental research, there have been no generic models to classify tumours and identify new types of cancer. Similarly, advances in the molecular classification of tumours may play a central role in cancer treatment. In this paper, a new approach to genome expression pattern interpretation is described and applied to the recognition of B-cell malignancies as a test set. Using DNA microarray data generated by A.A. Alizadeh et al. (2000), a neural network model known as a simplified fuzzy ARTMAP is able to identify normal and diffuse large B-cell lymphoma (DLBCL) patients. Furthermore, it discovers the distinction between patients with molecularly distinct forms of DLBCL without previous knowledge of those subtypes

Keywords

ART neural nets; DNA; arrays; cancer; data mining; fuzzy neural nets; genetics; learning (artificial intelligence); medical computing; pattern classification; tumours; B-cell malignancy recognition; DNA microarray data; cancer class discovery; cancer class prediction; cancer treatment; connectionist learning approach; diffuse large B-cell lymphoma; genome expression pattern interpretation; molecular classification; molecularly distinct forms; neural network model; simplified fuzzy ARTMAP; tumour classification; Bioinformatics; Biological information theory; Cancer; DNA; Gene expression; Genomics; Medical treatment; Monitoring; Organisms; Tumors;

fLanguage

English

Publisher

ieee

Conference_Titel

Bio-Informatics and Biomedical Engineering, 2000. Proceedings. IEEE International Symposium on

Conference_Location

Arlington, VA

Print_ISBN

0-7695-0862-6

Type

conf

DOI

10.1109/BIBE.2000.889609

Filename

889609