Integrating text mining, data mining, and network analysis for identifying genetic breast cancer trends

Area of Science:

• Biomedical research
• Genomics
• Bioinformatics

Background:

• Breast cancer poses a significant global health challenge, necessitating the identification of reliable genetic biomarkers.
• The exponential growth of scientific literature on breast cancer complicates manual identification of novel biomarkers.
• Existing literature contains valuable information for biomarker discovery, but efficient extraction methods are needed.

Purpose of the Study:

• To present a novel framework for discovering potential breast cancer genetic biomarkers.
• To integrate text mining and social network analysis for enhanced literature data investigation.
• To identify informative discoveries and trends in breast cancer research.

Main Methods:

• Utilized PubMed as the data source for analysis.
• Applied text mining techniques to extract relevant information from abstracts.
• Employed social network analysis to investigate gene-gene interactions and novel relationships (gene-year, gene-country, abstract-country).

Main Results:

• Identified gene-gene interactions and novel relationships over time and across countries.
• Revealed variations in discoveries and research group interests globally.
• Observed that different genes are highlighted in relation to specific countries, despite shared functionality.

Conclusions:

• The integrated framework successfully identifies potential breast cancer biomarkers from literature.
• Analysis revealed interesting trends in global research focus and gene associations.
• Text analysis results were validated against external tools for gene-gene relations and functions.