Optimisation strategies for directed evolution without sequencing.

Area of Science:

• Biotechnology
• Molecular Biology
• Synthetic Biology

Background:

• Directed evolution engineers biological systems by iterative mutagenesis and selection, often requiring DNA sequencing and synthesis.
• Current optimization methods for directed evolution leverage machine learning but are resource-intensive and incompatible with in vivo mutagenesis.
• Established techniques like Fluorescence-Activated Cell Sorting (FACS) offer sorting-based selection but can be limited by sequencing requirements.

Purpose of the Study:

• To explore optimization strategies for directed evolution that minimize or eliminate the need for DNA sequencing.
• To adapt these sequencing-independent methods for emerging single-cell selection techniques based on multiple measurable traits.
• To demonstrate the efficacy of these alternative strategies on established empirical landscapes.

Main Methods:

• Developed and explored optimization approaches for directed evolution within the constraints of sorting-based techniques like FACS.
• Expanded methods to accommodate emerging directed evolution techniques enabling single-cell selection based on combined measurable traits.
• Applied and validated strategies on the GB1 and TrpB empirical landscapes.

Main Results:

• Demonstrated potential for directed evolution optimization without DNA sequencing or synthesis.
• Showcased adaptability to advanced single-cell selection methods.
• Achieved up to 19-fold and 7-fold increases in the probability of reaching the global fitness peak on GB1 and TrpB landscapes, respectively.

Conclusions:

• Optimized directed evolution strategies can be implemented without sequencing, reducing resource intensity.
• These methods are compatible with emerging single-cell selection technologies, broadening applicability.
• The developed strategies significantly enhance the efficiency of achieving desired functional outcomes in biological engineering.