Gene Expression Data Analysis Using Feature Weighted Robust Fuzzy c-Means Clustering.

Area of Science:

• Bioinformatics
• Computational Biology
• Data Mining

Background:

• Gene expression data clustering is vital for understanding gene function and regulation.
• Features (genes) have varying relevance and redundancy in clustering tasks.
• Existing methods may not optimally account for feature contributions.

Purpose of the Study:

• To propose a novel fuzzy c-means clustering approach that considers feature contributions.
• To enhance the accuracy and relevance of clusters in gene expression data.
• To improve the understanding of gene functions and regulation through improved clustering.

Main Methods:

• Modified the fuzzy c-means objective function using weighted Euclidean distance.
• Incorporated a monotonically decreasing function to control feature contributions.
• Validated the approach using standard gene expression datasets and clustering performance measures.

Main Results:

• The proposed method demonstrates improved performance in partitioning data into relevant clusters.
• Weighted feature contributions lead to more biologically meaningful groupings.
• Performance measures show superiority over existing state-of-the-art clustering methods.

Conclusions:

• The novel fuzzy c-means approach effectively handles feature relevance in gene expression data.
• This method offers a more refined way to analyze gene expression patterns.
• It provides a valuable tool for biological data mining and discovery.