Graph-based unsupervised feature selection and multiview clustering for microarray data.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Microarray experiments generate large volumes of gene expression data requiring sophisticated analysis.
• Identifying informative genes is crucial for extracting biologically meaningful information and aiding disease diagnosis, prognosis, and treatment.

Purpose of the Study:

• To present an unsupervised feature selection technique for explorative data analysis of high-dimensional gene expression data.
• To address the challenge of extracting multiple, diverse clustering views from complex datasets.

Main Methods:

• Developed an unsupervised feature selection technique.
• Focused on extracting multiple clustering views that capture data diversity.
• Evaluated the technique on benchmark datasets.

Main Results:

• The proposed model demonstrates potential and effectiveness in analyzing gene expression data.
• Experimental results show favorable comparisons against traditional single-view clustering models.
• The technique outperforms other existing methods for the studied datasets.

Conclusions:

• The unsupervised feature selection technique offers a powerful approach for analyzing high-throughput gene expression data.
• Extracting multiple clustering views enhances the ability to uncover hidden biological information.
• This method aids in advancing the understanding and application of microarray data in disease research.