Employing social network analysis for disease biomarker detection

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Biomarker detection is crucial for disease diagnosis, especially for cancers.
• In silico genomic experiments are vital for analyzing large datasets.
• Current methods for biomarker discovery can be complex and data-intensive.

Purpose of the Study:

• To present a new approach for detecting cancer biomarkers using genomic microarray data.
• To leverage Social Network Analysis (SNA) techniques for biomarker identification.
• To reduce the number of features (genes) for efficient analysis and further biological research.

Main Methods:

• Utilizing Social Network Analysis (SNA) to model gene relationships.
• Representing genes as actors in a social network, with similarities as connections.
• Applying SNA to genomic microarray data for biomarker discovery.

Main Results:

• The SNA-based approach effectively identifies cancer biomarkers.
• This method significantly reduces the number of features from large genomic datasets.
• Classification performance using selected biomarkers is comparable or superior to using the entire dataset.

Conclusions:

• Social Network Analysis offers a promising and efficient strategy for cancer biomarker detection.
• This approach can streamline biological research by minimizing the number of candidate biomarkers.
• The method has the potential to reduce the need for extensive in vitro experiments.