biotextgraph: graphical summarization of functional similarities from textual information

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Functional interpretation of biological entities, like differentially expressed genes, is crucial in bioinformatics.
• Existing enrichment analysis (EA) methods primarily rely on biological pathway databases.
• Textual descriptions of biological entities in public databases are underutilized for uncovering novel biological mechanisms.

Purpose of the Study:

• To present biotextgraph, a novel R package for graphical summarization of omics textual data.
• To enable the assessment of functional similarities among lists of biological entities using text mining.
• To complement traditional EA by integrating textual information for deeper biological insights.

Main Methods:

• Development of the biotextgraph R package.
• Application of graphical summarization to omics textual description data.
• Comparative analysis with traditional enrichment analysis.

Main Results:

• The biotextgraph package provides graphical visualizations of textual data associated with biological entities.
• Text-based analysis revealed biologically meaningful terms not identified by pathway databases alone.
• The package facilitates the annotation of gene identifiers and integrates with EA.

Conclusions:

• Biotextgraph offers a valuable approach for routine omics data analysis by leveraging textual descriptions.
• Textual data visualization can uncover novel biological mechanisms missed by pathway-centric methods.
• The package enhances the functional interpretation of omics data, complementing existing bioinformatics tools.