Gene Ontology based housekeeping gene selection for RNA-seq normalization.

Area of Science:

• Bioinformatics
• Genomics
• Molecular Biology

Background:

• RNA-sequencing (RNA-seq) analysis is vital for understanding gene expression and protein function.
• Accurate normalization across diverse RNA-seq datasets is challenging but essential for identifying differentially expressed genes.
• Existing normalization methods may not adequately address inter-sample variability in complex experimental designs.

Purpose of the Study:

• To develop a novel computational method for selecting optimal housekeeping genes for RNA-seq inter-sample normalization.
• To enhance the accuracy and reliability of differential gene expression analysis.
• To provide biologists with a robust tool for calibrating multiple experimental datasets.

Main Methods:

• Utilizing user-defined keywords, Gene Ontology (GO) annotations, and GO term distance matrices.
• Incorporating orthologous housekeeping gene candidates and stability ranking based on coefficient of variation.
• Identifying functionally irrelevant housekeeping genes by selecting those with the most distant GO terms from experimental keywords and low inter-sample variation.

Main Results:

• Demonstrated that housekeeping gene selection significantly impacts differential gene expression analysis outcomes.
• Showcased the proposed method's superior performance compared to traditional approaches in both sensitivity and specificity.
• Validated the method's robustness and accuracy on novel and benchmark RNA-seq datasets.

Conclusions:

• The proposed method offers a robust and accurate approach for selecting housekeeping genes for inter-dataset normalization in RNA-seq studies.
• This strategy improves the reliability of differential gene expression analyses, particularly across varied experimental conditions.
• The tool facilitates more precise biological interpretation of gene expression data.