Global multiple protein-protein interaction network alignment by combining pairwise network alignments.

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Growing volume of protein-protein interaction network (PPIN) data necessitates advanced analysis techniques.
• Comparing and aligning PPINs across species (PPIN alignment) offers insights into evolution, gene function, and disease mechanisms.
• Existing methods for multiple network alignment (MNA) are limited, despite the analytical potential of aligning multiple PPINs simultaneously.

Purpose of the Study:

• To develop a novel algorithm for multiple protein-protein interaction network alignment (MNA).
• To enable the creation of MNAs from existing pairwise alignment tools, addressing limitations in native MNA implementations.

Main Methods:

• Propose SMAL, a scaffold-based MNA algorithm.
• SMAL converts results from any global pairwise alignment algorithm into an MNA in linear time.
• Utilize publicly available pairwise network aligners to construct multiple network alignments.

Main Results:

• SMAL successfully generated multiple network alignments using PPINs from eight species.
• Performance evaluation demonstrated SMAL's effectiveness in aligning up to eight PPINs.
• Scaffold network choice was examined for its effect on alignment quality.

Conclusions:

• SMAL provides a flexible approach to MNA, leveraging existing pairwise alignment tools.
• SMAL's performance is comparable to established MNA methods like IsoRankN and SMETANA.
• SMAL significantly outperforms other methods in computational time while preserving key alignment characteristics.