Alpha-plane based automatic general type-2 fuzzy clustering based on simulated annealing meta-heuristic algorithm for...

Area of Science:

• Bioinformatics
• Computational Biology
• Data Science

Background:

• Microarray gene expression data analysis is crucial for understanding biological processes.
• Clustering algorithms are essential for identifying patterns in high-dimensional biological data.
• Handling uncertainty in gene expression data remains a significant challenge.

Purpose of the Study:

• To develop a robust two-stage meta-heuristic algorithm for clustering gene expression data.
• To address the challenge of high uncertainty in biological datasets.
• To improve the performance of clustering in complex biological environments.

Main Methods:

• A novel objective function utilizing α-planes for general type-2 fuzzy c-means clustering was developed.
• A two-stage optimization algorithm integrating Simulated Annealing with a heuristic local search was proposed.
• The algorithm incorporates annealing, perturbation mechanisms, and iterative refinement for robust clustering.

Main Results:

• The proposed algorithm demonstrated superior performance in clustering synthesized and real-world microarray gene expression datasets.
• Experimental results showed significant improvements compared to existing state-of-the-art clustering techniques.
• The approach effectively handles high uncertainty, leading to more reliable data partitioning.

Conclusions:

• The novel two-stage meta-heuristic algorithm offers a powerful and effective solution for gene expression data clustering.
• The integration of general type-2 fuzzy sets and Simulated Annealing enhances robustness in uncertain environments.
• This method provides a valuable tool for bioinformatics research and discovery.