| | | |

Health Initiative

Intelligent System for the Classification of Multiple Tumour Types Using Gene Expression Data from DNA Microarray

The background that brings the necessity of this project is the following:

Current clinical diagnosis of human malignancies depends on a variety of morphological and histopathological data, and poses many challenges due to the subjective elements of the analyses, morphological heterogeneity of tumours, and the presence of many atypical tumours that do not fit well within currently established subtypes. Expression genomics approaches using DNA microarrays show promise of greatly reducing these challenges, by providing non-subjective data, and thus allowing for conformation and also development of more precise subtypes. The proposed project deals with the development of novel methods for tumour classification and use of these methods for the development of web-based “intelligent” software packages that will assist in the more accurate diagnosis and prognosis of tumour subtypes.

The aims of this project are the following:

Develop new intelligent methods for the automated diagnosis and prognosis of tumour types from gene expression data generated by DNA microarrays.
Develop the web-based “intelligent” software packages that will assist in the more accurate diagnosis and prognosis of tumour subtypes from gene expression data generated by DNA microarrays.

The possible impacts of our research project are as following:

In addition to purely scientific interest, we believe that this research proposal will be of a great importance in adapting and developing new robust multiple criteria classification methods. These methods will help to classify not only different type of tumours but also non-medical classification problems. We also believe that our web application will be used as a model for other bioinformatics laboratories that need to develop web-based tools in a short amount of time and on a limited budget.

This project started in August 2003 and is expected to end in July 2007.

We expect our research to contribute to the following:

Incorporate gene expression profiles and histopathological data, using novel methods in order to designate tumour classes
Select a subset of genes capable of discriminating the newly designated tumour classes, and create a low density DNA microarray based on this gene set
Adapt and apply the newly developed “fuzzy classification methods” to tumour diagnosis and prognosis using gene expression
Develop classification methods and determine set of discriminative genes that will be used for the development of low density chip and automatic web server that can be used in clinical practice

We expect our research to contribute to the following domains:

Data mining
Fuzzy Classification
Clinical Decision support system
Tumour diagnosis and prognosis from DNA microarray
Web-based intelligent software

Our research results could have an impact on the following sectors:

Oncology and pathology departments
Biochemistry
Pharmacists
Bioinformatics

The possible applications of this project include:

Molecular tumours diagnosis and prognosis from DNA microarray
Gene markers for tumours diagnosis
Cellular pathways
Gene functionalities

The possible technical transfer opportunities related to this project include:

Drug discovery
Tumours diagnosis and prognosis
Tumours treatment
Tumours prevention

If interested, please contact the Business Development Officer listed below.

The research partner of this project:

Beauséjour Medical Research Institute (BMRI), Moncton, New Brunswick

Related NRC-IIT Publications

Research Contact

Dr. Nabil Belacel
Research Officer
Health Initiative
NRC Institute for Information Technology
55 Crowley Farm Road, Suite 1100-C, Scientific Park
Moncton, NB E1A 7R1
Telephone: +1 (506) 861-0963
Fax: +1 (506) 851-3630
E-mail: Nabil.Belacel@nrc-cnrc.gc.ca

Business Contact

Georges Corriveau
Business Development Officer
Business Development Office, New Brunswick
NRC Institute for Information Technology
55 Crowley Farm Road, Suite 1100-C, Scientific Park
Moncton, NB E1A 7R1
Telephone: +1 (506) 861-0954
Fax: +1 (506) 851-3630
E-mail: Georges.Corriveau@nrc-cnrc.gc.ca