We are happy to share the article Simulated Design-Build-Test-Learn Cycles for Consistent Comparison of Machine Learning Methods in Metabolic Engineering as part of the special issue "AI for Synthetic Biology" from ACS Synthetic Biology.
Design-Build-Test-Learn Cycles are widely used to optimize microorganisms in an iterative fashion to increase their product yield. DBTL cycles strategies that find the best-producing strain, while minimizing the experimental effort remains an open question due to the time-consuming and costly nature of these experiments. This makes comparisons between different library transformations, noise scenarios, and machine learning models practically infeasible. To overcome this challenge, we introduce a mechanistic kinetic model-based framework to test and optimize machine learning for iterative combinatorial pathway optimization.
Feel free to reach out to us if you have any questions about the work.