Gene selection for microarray data analysis using principal component analysis

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Principal Component Analysis (PCA) is vital for multivariate data analysis and dimensionality reduction.
• PCA is a valuable tool in microarray data analysis, simplifying complex gene expression datasets.
• Comparing gene expression differences and classifying samples with numerous genes presents challenges.

Purpose of the Study:

• To propose a novel gene selection method for microarray data analysis.
• To enhance the effectiveness of Principal Component Analysis (PCA) in handling high-dimensional gene expression data.
• To identify the optimal gene subset for preserving the original data structure in cancer gene expression studies.

Main Methods:

• A gene selection method based on Krzanowski's strategy was developed.
• The proposed method was applied to a cancer gene expression dataset.
• Performance was evaluated by comparing with existing gene selection strategies.

Main Results:

• The proposed gene selection method demonstrated effectiveness in dimensionality reduction.
• The method successfully preserved the original data structure.
• It outperformed several other gene selection strategies in preserving data integrity.

Conclusions:

• The proposed gene selection method is effective for microarray data analysis.
• This approach offers a superior way to select gene subsets for preserving data structure.
• It provides a robust tool for analyzing complex cancer gene expression data.