Improving the organization and interactivity of metabolic pathfinding with precomputed pathways

Area of Science:

• Metabolic engineering and synthetic biology.
• Computational biology and bioinformatics.

Background:

• Growing knowledge of metabolic processes enables combining pathways from different species for chemical production.
• Current computational search algorithms for heterologous pathways yield numerous results, often repeating common subpaths.
• A need exists to reduce computational resources by leveraging recurring metabolic subpath modules.

Purpose of the Study:

• To develop and evaluate a meta-algorithm, Hub Pathway search with Atom Tracking (HPAT), for efficient metabolic pathfinding.
• To utilize a precomputed network of common hub metabolites and their connections to streamline pathway searches.
• To create an interactive webserver for visualizing metabolic pathways.

Main Methods:

• Developed HPAT, a meta-algorithm incorporating a precomputed network of subpath modules.
• Created a hub metabolite network table to guide searches.
• Implemented an interactive webserver for pathway visualization and filtering.

Main Results:

• HPAT identified 11 out of 19 known pathways, outperforming a comparable algorithm that found only 7.
• HPAT successfully identified novel 3-hydroxypropanoate (3-HP) synthesis pathways.
• Interactive filters reduced pathway visualization from hundreds to fewer than ten results, enhancing clarity.

Conclusions:

• This study introduces the first integration of a precomputed subpath network into metabolic pathfinding.
• HPAT demonstrates efficient discovery of alternate metabolic pathways through modularity.
• The developed method provides concise and interactive visualization of pathway search results.