A novel feature selection algorithm for identifying hub genes in lung cancer

Area of Science:

• Biomedical Informatics
• Computational Biology
• Genomics

Background:

• Lung cancer remains a leading cause of mortality globally, necessitating advanced diagnostic and therapeutic strategies.
• Accurate identification of cancer biomarkers is crucial for effective clinical management.
• Existing methods for gene expression analysis and feature selection require enhancement for complex diseases like lung cancer.

Purpose of the Study:

• To introduce and evaluate the Voting-Based Enhanced Binary Ebola Optimization Search Algorithm (VBEOSA) for feature selection in lung cancer research.
• To identify key hub genes and molecular pathways associated with lung cancer using VBEOSA.
• To explore the potential of VBEOSA in improving lung cancer diagnosis and understanding its molecular mechanisms.

Main Methods:

• Application of the Voting-Based Enhanced Binary Ebola Optimization Search Algorithm (VBEOSA), an ensemble method combining binary optimization and Ebola optimization search algorithm.
• Utilizing a lung cancer gene expression dataset from The Cancer Genome Atlas (TCGA) with preprocessing including outlier detection, normalization, and filtration.
• Performing protein-protein interaction analysis, pathway analysis, and weighted gene co-expression network analysis (WGCNA).

Main Results:

• VBEOSA successfully performed feature selection on the lung cancer dataset, identifying ten significant hub genes: ADRB2, ACTB, ARRB2, GNGT2, ADRB1, ACTG1, ACACA, ATP5A1, ADCY9, and ADRA1B.
• Pathway analysis highlighted the involvement of pathways such as salivary secretion and calcium signaling in lung cancer.
• WGCNA identified gene modules strongly correlated with clinical attributes of lung cancer.

Conclusions:

• VBEOSA demonstrates efficacy in feature selection for identifying lung cancer-associated hub genes and pathways.
• The identified genes and pathways offer valuable insights into the molecular landscape of lung cancer.
• This research provides a foundation for enhanced diagnostic capabilities and clinical management of lung cancer.