graphite - a Bioconductor package to convert pathway topology to gene network

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Gene set analysis increasingly incorporates pathway topology.
• Pathway elements include complex entities like protein complexes and gene families.
• Current R/Bioconductor packages lack comprehensive pathway network reconstruction, failing to handle complex entities or propagate signals through chemical compounds.

Purpose of the Study:

• To present graphite, a Bioconductor package for reconstructing gene-gene networks from multiple pathway databases.
• To address limitations in existing packages regarding complex pathway elements and signal propagation.

Main Methods:

• Interpreting pathway information from four databases using biologically-driven rules.
• Reconstructing gene-gene networks that incorporate protein complexes and gene families.
• Removing chemical compounds from the final graphs to simplify network representation.

Main Results:

• graphite provides a uniform resource for pathway analyses by generating integrated gene-gene networks.
• The package accounts for protein complexes and gene families, improving topological representation.
• It enables easy access to three recently proposed topological methods for pathway analysis.

Conclusions:

• graphite is an innovative Bioconductor package that consolidates information from major pathway databases.
• It simplifies pathway complexity by considering the biological meaning of pathway elements.
• The package serves as a valuable resource for topological pathway analysis.