Designing gene manipulation schedules for high throughput parallel construction of objective strains

Area of Science:

• Synthetic Biology
• Bioengineering
• Computational Biology

Background:

• Biofoundries accelerate strain development via parallel construction and design-build-test-learn (DBTL) cycles.
• Iterative gene manipulation for large-scale strain construction is time-consuming and expensive, hindering commercialization.

Purpose of the Study:

• To develop and validate a computational method for optimizing genetic manipulation schedules in biofoundries.
• To reduce the cost and time associated with constructing large numbers of engineered strains.

Main Methods:

• Introduced two complementary algorithms: greedy search of common ancestor strains (GSCAS) and minimizing total manipulations (MTM).
• GSCAS identifies and clusters common ancestor strains to reduce the number of unique constructions.
• MTM further minimizes genetic manipulations within the optimized schedule.

Main Results:

• A case study with 94 target strains showed GSCAS reduced gene manipulations by 36%, and MTM reduced them by an additional 10%.
• The method demonstrated robust performance across various gene manipulation frequencies.
• The approach effectively reduces the total number of strains to be constructed, creating a tree-like descendant structure.

Conclusions:

• The GSCAS-MTM method significantly enhances cost-efficiency and accelerates the development of commercial strains.
• Optimized genetic manipulation scheduling is crucial for efficient biofoundry operations.
• The developed algorithms are publicly available for biofoundry applications.