- Title
- Consensus clustering of gene expression microarray data using genetic algorithms
- Creator
- Mendes, Alexandre
- Relation
- Third IAPR International Conference on Pattern Recognition in Bioinformatics (PRIB 2008). PRIB 2008 Supplementary Proceedings (Melbourne, Vic. 15-17 October, 2008) p. 1-12
- Publisher Link
- http://dx.doi.org/10.1007/978-3-540-88436-1
- Publisher
- Gippsland School of Information Technology, Monash University
- Resource Type
- conference paper
- Date
- 2008
- Description
- This work presents a new consensus clustering method for gene expression microarray data based on a genetic algorithm. Using two datasets - DA and DB - as input, the genetic algorithm examines putative partitions for the samples in DA, selecting biomarkers that support such partitions. The biomarkers are then used to build a classifier which is used in DB to determine its samples classes. The genetic algorithm is guided by an objective function that takes into account the accuracy of classification in both datasets, the number of biomarkers that support the partition, and the distribution of the samples across the classes for each dataset. To illustrate the method, two whole-genome breast cancer instances from different sources were used. In this application, the results indicate that the method could be used to find unknown subtypes of diseases supported by biomarkers presenting similar gene expression profiles across platforms. Moreover, even though this initial study was restricted to two datasets and two classes, the method can be easily extended to consider both more datasets and classes.
- Subject
- microarray technologies; gene expression; genetic algorithms
- Identifier
- uon:6201
- Identifier
- http://hdl.handle.net/1959.13/802779
- Identifier
- ISBN:9780732622268
- Reviewed
- Hits: 720
- Visitors: 934
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|